Visit Rakiura

David Kipping

Assistant Professor, Columbia University

Full coverage. Every paper by this person that we could obtain has been read.

Affiliations

  • Assistant Professor, Columbia University (since 2015)
  • Menzel Fellow, Harvard-Smithsonian Center for Astrophysics (2014 to 2015)
  • Sagan Fellow, Harvard-Smithsonian Center for Astrophysics (2011 to 2015)
  • Postdoctoral fellow, Harvard-Smithsonian Center for Astrophysics (2011 to 2011)

Skills

Confirmed in multiple papers

?

Supported by quotes from two or more independent papers, or a contributions statement naming this person.

Lomb-Scargle periodogramUsed

An algorithm for detecting and characterizing periodic signals in unevenly-sampled data. The Lomb-Scargle periodogram has a particularly wide use within the astronomy community. This method allows efficient computation of a Fourier-like power spectrum estimator from such unevenly-sampled data, resulting in an intuitive means of determining the period of oscillation.

Supported by 2 papers.

Show the supporting quotes
Lomb-Scargle periodogram of the out-of-transit normalized flux of KOI-303 from quarters 1-11 and quarter 13.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI
Lomb-Scargle periodogram of each available \ quarter for KIC-5351250/Kepler-150.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

Bayesian evidenceUsed

Supported by 3 papers.

Show the supporting quotes
By comparing the Bayesian evidence of a planet-only model to that of a planet-with-moon, the detection process is handled in a Bayesian framework.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI
Light curve fits are performed in a Bayesian framework with the goal of both deriving parameter posteriors and computing the Bayesian evidence ( Z ) for each model.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI
Using the Bayesian evidence computed by , we find a planet+moon+TTVs model is favored over a planet+TTVs model with a formal significance of 5.2 \, ( = 15.3 0.2 ).
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

Bayesian model selectionUsed

Supported by 2 papers.

Show the supporting quotes
A planet+moon model has significantly greater freedom than a planet-only case guaranteeing a higher maximum likelihood, thereby necessitating the use of Bayesian model selection.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI
By using this code, we not only obtain parameter posteriors but also the Bayesian evidence of each model attempted, thus allowing for later Bayesian model selection and model averaged posteriors.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Cosine Filtering with Autocorrelation MinimizationUsed

Supported by 2 papers.

Show the supporting quotes
Long-term variations, such as focus drift and rotational modulations, are detrended using the Cosine Filtering with Autocorrelation Minimization ( ) algorithm.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI
In this work, we utilize the \ (Cosine Filtering with Autocorrelation Minimization) algorithm described in for detrending these various effects.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Durbin-Watson statisticUsed

Supported by 2 papers.

Show the supporting quotes
Our objective is therefore to choose a value of x such that the Durbin-Watson statistic is consistent with the lowest quantity of autocorrelation, when evaluated on the data surrounding the planetary transit.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI
However, \ is able to attempt dozens of different harmonics and evaluate the autocorrelation at a pre-selected timescale (we use 30\,minutes) and then select the harmonic order which minimizes this autocorrelation, as quantified using the Durbin-Watson statistic.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

exponential distributionUsed

Supported by 3 papers.

Show the supporting quotes
When we consider that this CDF describes the probability that life arose by a time t , it therefore follows that the probability distribution for the time at which life first arose, which we can denote as , must be an exponential distribution i.e. t_life ( ) .
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI
All cases use the same inference model, an exponential distribution, and show 20 lines for the 20 MCMC walkers used.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI
Each then experiences a constant hazard rate of collapse, which defines an exponential distribution with rate parameter .
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

median absolute deviationUsed

Supported by 2 papers.

Show the supporting quotes
Dividing the fluxes by their formal reported uncertainties, we find that the robust r.m.s. (given by 1.4286 multiplied by the median absolute deviation) equals 1.09, indicating only a small amount of extra noise above the reported uncertainties.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI
These light curves were then detrended using the same \ algorithm as utilized on the simulated data, normalizing, and rejecting once again badly detrended light curves by evaluating the median absolute deviation.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

Monte Carlo simulationsUsed

Supported by 2 papers.

Show the supporting quotes
Points represent the median from 10^4 Monte Carlo simulations, the errors represent the 68.3% inter-quantile range and the solid line is a sigmoid fit.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI
I then generated 10^6 Monte Carlo simulations where all of the parameters are randomly varied and four versions of the term in the modulus are calculated each time, for each solution for f_S,max .
The Transits of Extrasolar Planets with Moons (2011) DOI

multimodal nested sampling algorithmUsed

Supported by 2 papers.

Show the supporting quotes
Selected targets are fitted using a multimodal nested sampling algorithm coupled with a planet-with-moon light curve modelling routine.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI
Model light curves are fitted to the data using the multimodal nested sampling algorithm \ .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

photoeccentric effectUsed

Supported by 2 papers.

Show the supporting quotes
Next, we compare the transit derived stellar density (which assumes a circular orbit) to an independent estimate, in order to exploit the photoeccentric effect to infer a minimum eccentricity, e_min , of each KOI .
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI
We identify a sample of 17 transiting exoplanets around late-type stars with similar radii and irradiation to that of Earth and use the ``photoeccentric effect'' - which exploits transit durations - to infer their eccentricity distribution via hierarchical Bayesian modelling.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

quadratic limb darkening lawUsed

Supported by 2 papers.

Show the supporting quotes
We assume a quadratic limb darkening law for the host star, where the specific intensity is described by I( ) = 1 - u_1 (1- ) - u_2 (1- )^2 where =1-r^2 , r is the normalized radial coordinate on the stellar disk and u_1 & u_2 are the limb darkening coefficients.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI
We use a quadratic limb darkening law, interpolating the WD DA Z=0 table of coefficients (Table 100C) for g = 7.915 and T_eff = 4710 (best reported values from ) to give u_1 = 0.0101 and u_2 = 0.4091 .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

Simple Aperture Photometry (SAP)Used

Supported by 3 papers.

Show the supporting quotes
The first step is simply to download the \ Simple Aperture Photometry (SAP) for each target.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI
As with previous papers, we only use the Simple Aperture Photometry (SAP) processed data in our analysis, rather than the Presearch Data Conditioning (PDCSAP) time series ( , ).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI
In what follows, we use the Simple Aperture Photometry (SAP) for quarters 1-15 and perform our own detrending rather than relying on the Presearch Data Conditioning (PDC) data.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

Supported by one paper

?

Supported by a quote from a single paper with the person in an attributable author position.

6th-degree polynomial fit to mass vs. ageUsed

Supported by 1 paper.

Show the supporting quotes
M is calculated with the derivative of a 6th-degree polynomial fit to mass vs. age and shown in units of M_ Ceres yr ^-1 , where M_ Ceres = 9.1 10^20\, kg.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

triple-layer networkUsed

Supported by 1 paper.

Show the supporting quotes
As a final test, we trained a triple-layer network with U_1=10 , U_2=8 and U_3=6 and similarly found R_0=1.70 .
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Nested samplingUsed

Monte Carlo method that integrates the Bayesian evidence by sampling within nested likelihood contours, yielding posterior samples as a by-product.

Supported by 1 paper.

Show the supporting quotes
One of the most efficient algorithms that satisfies these requirements is \ , based on nested sampling , and this is the method employed in this work and in our previous HEK papers.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

Poisson distributionUsed

A probability function that characterizes discrete random events occurring independently of one another within some definite time or space. It may be regarded as an approximation of the binomial distribution when the number of events becomes large and the probability of success becomes small. The Poisson distribution is expressed by: f(x) = (λxe-λ)/x!, where λ is the mean number of successes in the interval, e is the base of the natural logarithm, and x is the number of successes we are interested in.

Supported by 1 paper.

Show the supporting quotes
Under these assumptions, I show that N in the Drake Equation must follow another Poisson distribution, with a mean rate ( / ) .
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

F testUsed

UAT concept: F test (Parametric hypothesis tests)

Supported by 1 paper.

Show the supporting quotes
A modern statistical analysis using an F-test finds the three-body model to be accepted with 1.9- confidence, below the standard detection threshold.
The Transits of Extrasolar Planets with Moons (2011) DOI

Cross-validation Used

UAT concept: Cross-validation (Astrostatistics techniques)

Supported by 1 paper.

Show the supporting quotes
Cross-validation is performed by only training on 75% of the 1786 training examples comprising the \ input pattern (see Section for details). The remaining 446 training examples are treated as the validation set.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Bayes factorUsed

UAT concept: Bayes factor (Astrostatistics tools)

Supported by 1 paper.

Show the supporting quotes
A: Bayes factor between a model where life emerges rapidly ( _L 1/t_L' ) versus slowly ( _L 1/t_L' ) on the Earth.
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

Markov chain Monte Carlo samplingUsed

Drawing samples from a posterior distribution by constructing a Markov chain whose stationary distribution is the target.

Supported by 1 paper.

Show the supporting quotes
A] Log-likelihoods from the 20 independent MCMC walkers (coloured blue to yellow), showing good mixing.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

1 minute numerical re-samplingUsed

Supported by 1 paper.

Show the supporting quotes
The model curve is generated at precisely the same cadence as the data used for the planet-stack and accounts for the long-cadence integration time using 1 minute numerical re-sampling .
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

2-spot modelUsed

Supported by 1 paper.

Show the supporting quotes
Whilst the planet-moon model is clearly a better fit than the planet-only model, a 2-spot model is able to out-perform either and is well-motivated from the activity levels observed in the out-of-transit light curve.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

2nd and 1st order Hermite interpolationUsed

Supported by 1 paper.

Show the supporting quotes
Near the edges of the grid, we switch to 2nd and 1st order Hermite interpolation as appropriate.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

95% confidence upper limitsReviewed

Supported by 1 paper.

Show the supporting quotes
95% confidence upper limits provided in the top-right.
In Search of Exomoons (2014) DOI

a-posteriori credible intervalsUsed

Supported by 1 paper.

Show the supporting quotes
The joint posterior (bottom-left) naturally has a similar shape to the likelihood (Figure ) given our diffuse prior, but the contours now represent the a-posteriori credible intervals rather than likelihood confidence limits.
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

agreement metricUsed

Supported by 1 paper.

Show the supporting quotes
Only 177 of these transits (0.3%) met our highest standard of confidence, with a agreement metric greater than 0.9. This metric therefore provides additional discrimination power by about a factor of 5.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

algebraic probability of being alone in the galaxyUsed

Supported by 1 paper.

Show the supporting quotes
This is used to demonstrate why the Copernican Principle does not allow one to infer N , as well evaluating the algebraic probability of being alone in the galaxy.
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

AMOEBA routineUsed

Supported by 1 paper.

Show the supporting quotes
This approximate solution is then used as a starting point for a ^2 -minimisation performed with the AMOEBA routine .
The Transits of Extrasolar Planets with Moons (2011) DOI

analytic expressions for the times of transit minima and the transit durationIntroduced

Supported by 1 paper.

Show the supporting quotes
Chapter presents new analytic expressions for the times of transit minima and the transit duration, which will be critical in the later search for exomoons.
The Transits of Extrasolar Planets with Moons (2011) DOI

analytic expressions for the uncertainty on the time of transit minimumUsed

Supported by 1 paper.

Show the supporting quotes
To address this, I will use the analytic expressions for the uncertainty on the time of transit minimum ( _T ) and duration ( T ) as derived by using a Fisher-analysis of a trapezoid-approximated circular-orbit light curve.
The Transits of Extrasolar Planets with Moons (2011) DOI

artificial neural networks (ANNs)Used

Supported by 1 paper.

Show the supporting quotes
Using artificial neural networks (ANNs) trained on the \ catalog and making careful feature selection to account for the differing sensitivity of , we are able to predict the most likely short-period transiters to be accompanied by additional transiters.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

asymmetric Gaussian modelUsed

Supported by 1 paper.

Show the supporting quotes
Comparison of the original data presented in Figure 1 of for the binned occurence rate of exoplanets in the range 1-5\,AU (black), with our proposed asymmetric Gaussian model (red).
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

automatic target selection (TSA) algorithmUsed

Supported by 1 paper.

Show the supporting quotes
These seven candidates were identified through our automatic target selection (TSA) algorithm and target selection prioritization (TSP) filtering, whereby we excluded systems exhibiting significant time-correlated noise and focussed on those with a single transiting planet candidate of radius less than 6 \,R_ .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

Automatic Target Selection (TSA) algorithmExtended

Supported by 1 paper.

Show the supporting quotes
Assumed mass function of Kepler Objects of Interest (KOIs) used in the updated Automatic Target Selection (TSA) algorithm of this work.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

auxiliary transitsReviewed

Supported by 1 paper.

Show the supporting quotes
The former, which we refer to as ``auxiliary transits'', is more likely to be detected for moons on wide orbits.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

Bayesian evidencesUsed

Supported by 1 paper.

Show the supporting quotes
Bayesian evidences of various fits for KOI-722.01.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

Bayesian model averaging (BMA)Used

Supported by 1 paper.

Show the supporting quotes
BMA provides a coherent approach for including our uncertainty in the models themselves. To our knowledge, BMA has not been previously applied in exoplanet studies but we here introduce the first application.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Bayesian multimodal nested samplingUsed

Supported by 1 paper.

Show the supporting quotes
Using photodynamical modeling and Bayesian multimodal nested sampling, we find no compelling evidence for an exomoon in these eight systems.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

Bayesian photodynamicsUsed

Supported by 1 paper.

Show the supporting quotes
We present a survey of 41 \ Objects of Interest (KOIs) for exomoons using Bayesian photodynamics, more than tripling the number of KOIs surveyed with this technique.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

Bernoulli experimentsUsed

Supported by 1 paper.

Show the supporting quotes
For each set of N exoplanets, we draw a from its prior and perform N Bernoulli experiments with success probability p .
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

bi-linear interpolationUsed

Supported by 1 paper.

Show the supporting quotes
During the actual fits, we used a bi-linear interpolation of every unique binned photometric data point, conditioned upon the LUT.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

binary flagUsed

Supported by 1 paper.

Show the supporting quotes
In each training example, we define a binary flag, y_t=0 or y_t=1 , which describes whether there are additional transiting planets with P>P_cut (or ``outers'') .
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Binomial distributionUsed

Supported by 1 paper.

Show the supporting quotes
The probability of obtaining M successes from a total of N Bernouilli experiments describes a Binomial distribution and thus we may write that Pr(N,p;M) &= (1-p)^N-M p^M NM.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

blackbody curveUsed

Supported by 1 paper.

Show the supporting quotes
Using the two measurements at 8.0\, m and 24.0\, m (black dots with error bars), we fit a blackbody curve through the points (gray lines) with a nightside planetary temperature of T = 1148 32 K.
The Transits of Extrasolar Planets with Moons (2011) DOI

blending factorsUsed

Supported by 1 paper.

Show the supporting quotes
These blending factors are treated as Gaussian priors, with a standard deviation set by the uncertainty on the converted magnitude.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

bootstrap filterUsed

Supported by 1 paper.

Show the supporting quotes
In such a case, we are able to employ a highly efficient sampling algorithm known as the ``bootstrap filter'', which is a class of ``particle filter'' .
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

bootstrap of each pixel's light curve residualsUsed

Supported by 1 paper.

Show the supporting quotes
Centroid positions are provided at the top of each panel in Fig , with uncertainties estimated using a bootstrap of each pixel's light curve residuals.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

centroidsUsed

Supported by 1 paper.

Show the supporting quotes
In Fig. , we show the time series of the x (column) and y (row) centroids for Kepler-90 during Q6, after detrending with .
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

CHEOPS noise modelUsed

Supported by 1 paper.

Show the supporting quotes
Illustration of the noise model adopted in this work for CHEOPS, showing the different noise components affecting the total.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

chromatic cloakIntroduced

Supported by 1 paper.

Show the supporting quotes
A chromatic cloak, effective at all wavelengths, is more challenging requiring a large array of tunable lasers with a total power of 250\,MW.
A cloaking device for transiting planets (2016) DOI

cloak only the atmospheric signatures associated with biological activityIntroduced

Supported by 1 paper.

Show the supporting quotes
Alternatively, a civilization could cloak only the atmospheric signatures associated with biological activity on their world, such as oxygen, which is achievable with a peak laser power of just 160\,kW per transit.
A cloaking device for transiting planets (2016) DOI

cloaking through controlled laser emissionIntroduced

Supported by 1 paper.

Show the supporting quotes
We suggest that advanced civilizations could cloak their presence, or deliberately broadcast it, through controlled laser emission.
A cloaking device for transiting planets (2016) DOI

code developed for star liftingIntroduced

Supported by 1 paper.

Show the supporting quotes
The code developed for star lifting and data produced during the preparation of this manuscript is available at this github repository.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

combining TTV and TDVIntroduced

Supported by 1 paper.

Show the supporting quotes
Therefore, combining TTV and TDV together allows one to determine both the mass and period of the exomoon, which satisfies the first critical desideratum outlined in .
The Transits of Extrasolar Planets with Moons (2011) DOI

comparison of predicted planetary radii and reported measured planetary radiiUsed

Supported by 1 paper.

Show the supporting quotes
A comparison of predicted planetary radii from this work using the \ package of and reported measured planetary radii. Only nine planets in our sample have a measured radius from transits.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

Corner plotUsed

Supported by 1 paper.

Show the supporting quotes
Corner plot of our resulting posteriors for HD 20794c an illustrative example. Covariances are evident between the parameters, including between mass and radius.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

Cosine Filtering with Autocorrelation MinimizationIntroduced

Supported by 1 paper.

Show the supporting quotes
Data is detrended using a custom algorithm which we dub Cosine Filtering with Autocorrelation Minimization ( ), which is described in this section.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

Credible intervalsUsed

Supported by 1 paper.

Show the supporting quotes
Right: Credible intervals evaluated using 10^5 such samples as computed in the left panel.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

CSL.clsUsed

Supported by 1 paper.

Show the supporting quotes
This sample file uses a class file named CSL.cls where the authors should use during their manuscript preparation.
Do SETI optimists have a fine-tuning problem? (2025) DOI

Detrend LightcurveUsed

Supported by 1 paper.

Show the supporting quotes
A lightcurve to be detrended accompanied by the ``Detrend Lightcurve" dialog box. The green line is the fitted trend line. See Figure for the detrended result.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

difference image normalised by pixel-by-pixel uncertaintyUsed

Supported by 1 paper.

Show the supporting quotes
Right: Observed difference image normalised by pixel-by-pixel uncertainty.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

distorting their transit to an artificial shapeIntroduced

Supported by 1 paper.

Show the supporting quotes
Accordingly, we propose that a civilization may deliberately broadcast their technological capabilities by distorting their transit to an artificial shape, which serves as both a SETI beacon and a medium for data transmission.
A cloaking device for transiting planets (2016) DOI

diurnally average the solar fluxUsed

Supported by 1 paper.

Show the supporting quotes
We must diurnally average the solar flux: S = q_0 . This means we must first integrate over the sunlit part of the day, i.e. h=[-H, +H] , where H is the radian half-day length at a given latitude.
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

dual-layer networkUsed

Supported by 1 paper.

Show the supporting quotes
We also tried using a dual-layer network, exploring a variety of neuron combinations up to a maximum complexity of U_1=6 and U_2=6 , but find R_0 does not improve beyond 170 %.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Durbin-Watson metricUsed

Supported by 1 paper.

Show the supporting quotes
After every detrending, we compute the autocorrelation on a 30\,minute timescale using the Durbin-Watson metric:
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

dynamical TTV modelUsed

Supported by 1 paper.

Show the supporting quotes
Parameters come from both a transit light curve model and a dynamical TTV model to account for the planet-planet perturbations between KOI-314b/c and KOI-784.01/.02.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

early stoppingUsed

Supported by 1 paper.

Show the supporting quotes
One popular method for choosing the ANN architecture is known as early stopping, which we use in this work. Essentially, the objective is to start from a simple model and build up in complexity, stopping once the performance of cross-validation no longer improves.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

eccentric orbits for both the planet and the moonUsed

Supported by 1 paper.

Show the supporting quotes
Unlike previous surveys, we decided to investigate the potential for eccentric orbits for both the planet and the moon due to the high-priority nature of .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Empirical posterior distributionsReviewed

Supported by 1 paper.

Show the supporting quotes
Empirical posterior distributions for the mass ratio between a putative satellite and 17 Kepler planetary candidates surveyed by the HEK project.
In Search of Exomoons (2014) DOI

empirically calibrated probabilistic mass-radius relationUsed

Supported by 1 paper.

Show the supporting quotes
Using an empirically calibrated probabilistic mass-radius relation, \ , we address this by predicting a catalog of homogeneous credible intervals for these three keys terms for 468 planets discovered via radial velocities.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

ensemble of convolutional neural networks (CNNs)Used

Supported by 1 paper.

Show the supporting quotes
In this work, we train an ensemble of convolutional neural networks (CNNs) to identify candidate exomoon signals in single-transit events observed by .
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

epoch-by-epoch transit light curve detrendingUsed

Supported by 1 paper.

Show the supporting quotes
Transit light curves are detrended on an epoch-by-epoch basis.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

evolutionary tracksUsed

Supported by 1 paper.

Show the supporting quotes
We explore stars with initial mass M_0 such that 0.2 M_0/ M_ 1.2 with increments of M_0=0.05\, M_ and metallicity Z \ \ 0.01, 0.1, 1 in units of Z_ = 0.0122 . Exploring this range for both implementations results in a total of 126 engineered evolutionary tracks.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

explicit forward time, centre space finite difference algorithmUsed

Supported by 1 paper.

Show the supporting quotes
The diffusion equation is solved using a simple explicit forward time, centre space finite difference algorithm (as was done in ).
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

extrapolations of the moon modelUsed

Supported by 1 paper.

Show the supporting quotes
The last 5 panels show the Q10-13 data with 50 extrapolations (50 different shadings used) of the moon model overlaid (parameters randomly drawn from the joint posteriors), which exhibit poor predictive power.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

feature importance metricUsed

Supported by 1 paper.

Show the supporting quotes
The reduced ^2 deviance (computed using Poisson counting errors) of these fractions relative to the naive flat probability, which would be expected for irrelevant features, is shown in the top-right of each panel, which can be treated as a feature importance metric.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Feature mapsUsed

Supported by 1 paper.

Show the supporting quotes
Feature maps for six example simulations containing moon signals. Red data points indicate locations where the moon prediction is unchanged by replacing the light curve with a median filter. The moon prediction changes to a no moon prediction when the blue points are replaced by a median filter.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

feature selectionUsed

Supported by 1 paper.

Show the supporting quotes
We ultimately elect to use R_max as our radius-like feature and P(R_max) as our period-like feature since they are both amongst the strongest features via the ^2 metrics, but more importantly because they will generally always be available to us as features, for the following reasons.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

filtered against bad dataUsed

Supported by 1 paper.

Show the supporting quotes
Data comes from printouts pulled by Marc Abel around September 1985, and then filtered against bad data (e.g. clock errors, printer failures, etc) by co-author Robert Gray.
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

finite-mass, zero-radius moon modelUsed

Supported by 1 paper.

Show the supporting quotes
As a more useful test in these two instances, we define the null model to be a finite-mass, zero-radius moon model H_S,R0 . This model is able to reproduce the sinusoidal TTVs but clearly a real exomoon case should be favored over a model negating the moon's radius.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

fitted candidate TTV signalUsed

Supported by 1 paper.

Show the supporting quotes
In both cases, the blue line shows the fitted candidate TTV signal.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

flux based measurementUsed

Supported by 1 paper.

Show the supporting quotes
A flux based measurement. Here, the depth and size of a candidate signal is being measured.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

Flux weighted centroid time seriesUsed

Supported by 1 paper.

Show the supporting quotes
Flux weighted centroid time series of . As visible from the plots, which are labeled in the top-left corner of each panel, the centroids exhibit a small shift during the time of the transits of .
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

follow-up criteriaExtended

Supported by 1 paper.

Show the supporting quotes
One significant change in this work is that we now choose the 25% follow-up data to lie somewhere around the middle of our total available time series. This means that F2 now tests an interpolation of the model rather than an extrapolation.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

forecasterUsed

Supported by 1 paper.

Show the supporting quotes
We also included the slight TTV and TDV effects induced by the Galilean moons by including (M_S/M_P) , which was computed by using forecaster applied to the moon radius to predict a mass, and then using the physical planetary masses predicted in .
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

forecasting the planetary massUsed

Supported by 1 paper.

Show the supporting quotes
After forecasting the planetary mass, we expect it to be capable of maintaining at least a 0.3\, M_ exomoon for 5\,Gyr, and the TTV signal corresponds to a moon mass as low as 0.75 Lunar masses.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

formulae for a sail's accelerationExtended

Supported by 1 paper.

Show the supporting quotes
To address this, we present formulae for a sail's acceleration, first in response to a single photon and then extended to an ensemble.
Relativistic Light Sails (2017) DOI

four test casesUsed

Supported by 1 paper.

Show the supporting quotes
In the first, four test cases are run to investigate in detail the dependence of the exomoon climate on orbital direction, orbital inclination, and on the frequency of stellar eclipse by the host planet.
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

Galilean analog GLC modelsUsed

Supported by 1 paper.

Show the supporting quotes
In addition to fitting the entire sample with effective moon and Galilean analog GLC models, we also performed GLC fits on a number of physically motivated subsets.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

Gaussian likelihood functionUsed

Supported by 1 paper.

Show the supporting quotes
For both models, a Gaussian likelihood function is adopted in our fits and exomoons are assumed to have near-circular orbits.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

generating Gaussian-noise around the trendUsed

Supported by 1 paper.

Show the supporting quotes
To achieve this we computed a moving median trend line for the light curve, masking the neighbor's transit (also using transit timings from , where available), and generating Gaussian-noise around the trend based on the noise profile of the light curve (see Figure ).
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

grid of individual likelihood evaluationUsed

Supported by 1 paper.

Show the supporting quotes
Comparison of our grid of individual likelihood evaluation (left) and the emulator used for the MCMC using interpolation (right).
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

grid of SNRsUsed

Supported by 1 paper.

Show the supporting quotes
We thus decided the best approach was to calculate a grid of SNRs based on a broad range of input parameters, and then use this grid to train an interpolation model that would be far faster to evaluate than the SNR numerical integration each time.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

habitability indicesUsed

Supported by 1 paper.

Show the supporting quotes
We calculate habitability indices in the same manner as .
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

HBMUsed

Supported by 1 paper.

Show the supporting quotes
Using an HBM, we infer the implied power-law radius distribution conditioned upon WD\,1856\,b and arrive at a top-heavy distribution, such that 0.1 - 2 \, R_ planets are an order-of-magnitude less common than 2 - 20 \, R_ planets in the period range of 0.1-10\,days.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

Heaviside Theta functionUsed

Supported by 1 paper.

Show the supporting quotes
Instead, for the sake of simplicity, we will assume that the completeness function is a Heaviside Theta function with a trigger value at SNR =10 .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

Hermite cubic interpolatorUsed

Supported by 1 paper.

Show the supporting quotes
We train use a Hermite cubic interpolator on this grid.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

hierarchical Bayesian modellingUsed

Supported by 1 paper.

Show the supporting quotes
We identify a sample of 17 transiting exoplanets around late-type stars with similar radii and irradiation to that of Earth and use the ``photoeccentric effect'' - which exploits transit durations - to infer their eccentricity distribution via hierarchical Bayesian modelling.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

Hill sphereUsed

Supported by 1 paper.

Show the supporting quotes
In the case of Kepler-90g, a more useful constraint on the planet's density comes from demanding the moon orbit within the Hill sphere, which requires (M_PM_ ) > 3 ( (R_P/R_ ) (a_SP/R_P) (a/R_ ) )^3, where the bracketed terms on the right-hand side are all observables from our light curve model.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

Histograms of the SNR_hourUsed

Supported by 1 paper.

Show the supporting quotes
Histograms of the SNR_hour (in log space) for the three broad class of planets considered in this work.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

hybrid 2/3-feature ANNExtended

Supported by 1 paper.

Show the supporting quotes
Cross-validation results from our hybrid 2/3-feature ANN, similar to Figure except we only show the fitted sigmoids.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

IAU Working Definition of ExoplanetUsed

Supported by 1 paper.

Show the supporting quotes
In this commentary, it is shown that both of these companions are technically “planets” according to the existing IAU Working Definition of Exoplanet.
What Even Is an “Exomoon”? (2026) DOI

independent light curve analysesUsed

Supported by 1 paper.

Show the supporting quotes
Here, we search through published \ TTV data for such signals, applying a battery of significance and robustness checks, plus independent light curve analyses for candidate signals.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

injectingUsed

Supported by 1 paper.

Show the supporting quotes
To test this, we created a fake Q6 time series by taking the pixel-level light curves of Kepler-90g's transit, reducing the depth to equal that of Kepler-90g.01, and injecting it in place of the original Kepler-90g.01 transit for each pixel.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

injecting planet transits into each light curveUsed

Supported by 1 paper.

Show the supporting quotes
We then apply the CNN ensemble to light curves from 1880 Kepler Objects of Interest with periods >10 days ( 57,000 individual transits), and further test the accuracy of the CNN classifier by injecting planet transits into each light curve, thus quantifying the extent to which residual stellar activity may result in false positive classifications.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

injection-recoveriesUsed

Supported by 1 paper.

Show the supporting quotes
Comparison of our a-posteriori distribution for the mean eccentricity (green) versus two injection-recoveries. The black lines show the inference when injecting a purely circular population, which is indistinguishable from that obtain with the real data (green).
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

injection-recoveryUsed

Supported by 1 paper.

Show the supporting quotes
The signal of the exomoon candidate, -i, is a 4.8-sigma effect and is persistent across different instrumental detrending methods, with a 1 % false-positive probability via injection-recovery.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

instrumental detrending methodsUsed

Supported by 1 paper.

Show the supporting quotes
The signal of the exomoon candidate, -i, is a 4.8-sigma effect and is persistent across different instrumental detrending methods, with a 1 % false-positive probability via injection-recovery.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

inter-parameter correlation matricesUsed

Supported by 1 paper.

Show the supporting quotes
For several proposed light curve fitting parameter sets (left column), I calculate the inter-parameter correlation matrices in the examples of i) a hypothetical near-grazing hot-Jupiter on a circular orbit ii) a system similar to the eccentric planet HD 80606b.
The Transits of Extrasolar Planets with Moons (2011) DOI

interpolate across the evolved values of L_* (or F )Used

Supported by 1 paper.

Show the supporting quotes
We then interpolate across the evolved values of L_* (or F ) at our desired constant value to solve for the required mass_change, where a final evolution for this timestep is carried out with this value.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

inverse transit-like signalIntroduced

Supported by 1 paper.

Show the supporting quotes
In the case of the Earth, the planet could be cloaked by generating an inverse transit-like signal of peak power 60\,MW.
A cloaking device for transiting planets (2016) DOI

isoirradianceUsed

Supported by 1 paper.

Show the supporting quotes
So, for a given age, we solve for the mass such that F L_*M_*^2 is held constant. Though both implementations hold bolometric irradiance constant for a planet at any orbital radius, we call this implementation where the lifted mass is moved outside the orbital radius isoirradiance.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

isoluminosityUsed

Supported by 1 paper.

Show the supporting quotes
In this case, keeping flux on the planet constant is equivalent to keeping the luminosity of the host star constant and we call this implementation isoluminosity.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

joint likelihood functionUsed

Supported by 1 paper.

Show the supporting quotes
The joint likelihood function for the rate of abiogenesis, _L , and the rate of intelligence emergence, _I (for cases where T>2t_L'+t_I' ).
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

jointly fit both data sets using a light curve modelUsed

Supported by 1 paper.

Show the supporting quotes
We thus obtained the corrected GTC photometry (A. Vanderburg, priv. comm.), detrended the \ photometry ourselves using \ , and then jointly fit both data sets using a light curve model and \ .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

kernel density estimatorUsed

Supported by 1 paper.

Show the supporting quotes
Here, the relation is not parametric but drawn from a kernel density estimator.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

Kullback-Leibler divergence (KLD)Used

Supported by 1 paper.

Show the supporting quotes
To evaluate the difference in a quantitative way, we choose to use the Kullback-Leibler divergence (KLD), which is also known as relative entropy .
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

latent priors on the derived mean planetary and satellite densitiesUsed

Supported by 1 paper.

Show the supporting quotes
As in our previous paper , we also have latent priors on the derived mean planetary and satellite densities, _P and _S (see for how these may be calculated).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

LcReporterUsed

Supported by 1 paper.

Show the supporting quotes
The Regular Signals worksheet produced by LcReporter.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

LcTools software systemIntroduced

Supported by 1 paper.

Show the supporting quotes
To assist citizen scientists in processing those lightcurves, the LcTools software system was developed.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

LcViewerUsed

Supported by 1 paper.

Show the supporting quotes
TTVs_Kepler_Schmitt -- A supplemental TTV library manually created with LcViewer by the lead author for use with TTVs_Kepler_Mazeh.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

least squares power-law modelUsed

Supported by 1 paper.

Show the supporting quotes
We took the four Galilean moons in isolation and calibrated a least squares power-law model to it, giving (R/R_ ) N(6.95 + 0.27 (a_SP/R_P), 0.17) , where the standard deviation quoted is that resulting from the residuals of the best-fitting line.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

LEBM simulationsUsed

Supported by 1 paper.

Show the supporting quotes
The classification of the LEBM simulations carried out using the posterior parameter sets as input.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

light curves accounting for limb darkeningUsed

Supported by 1 paper.

Show the supporting quotes
I use the code to generate light curves accounting for limb darkening based upon a non-linear law from .
The Transits of Extrasolar Planets with Moons (2011) DOI

likelihood emulator using the Big Ear observing logsUsed

Supported by 1 paper.

Show the supporting quotes
Here, we employ a likelihood emulator using the Big Ear observing logs to infer the probable signal properties under this hypothesis.
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

linear ephemeris model fit and transit timing variation model fitUsed

Supported by 1 paper.

Show the supporting quotes
Our script then creates two directories for a linear ephemeris model fit, H_LIN , and a transit timing variation model fit, H_TTV , to be fed into \ .
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

linear fit through the out-of-transit dataUsed

Supported by 1 paper.

Show the supporting quotes
We clip out the data within 2 T_Hill and apply a linear fit through the out-of-transit data to remove any residual trend, which acts as a final normalization.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

linearly interpolate the G5 and K5 total noise modelsUsed

Supported by 1 paper.

Show the supporting quotes
We linearly interpolate the G5 and K5 total noise models as a function of effective temperature, except for temperatures exceeding these two extremes, beyond which we simply adopt the nearest neighbor model.
Forecasting the detectability of known radial velocity planets with the upcoming CHEOPS mission (2018) DOI

local autocorrelationUsed

Supported by 1 paper.

Show the supporting quotes
Each model is regressed to the data then ranked by the local autocorrelation, as computed using the statistic (DW) on the timescale of the LC cadence, with the lowest autocorrelation being favored.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

log Bayes factorUsed

Supported by 1 paper.

Show the supporting quotes
Histogram of the log Bayes factor between a planet-moon and planet-only model from 200 fake planet-only signals injected into the light curve.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

log uniform distributionUsed

Supported by 1 paper.

Show the supporting quotes
For this reason, the choice of prior is less crucial than before and we elect to adopt a simple objective prior in the form of a log uniform distribution because it is always positive, admits a wide range of order of magnitudes, and is non-informative with the maximum entropy (in log space): Pr( ) = 1 1 _max - _min.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

log-likelihood of observing the two transit timesUsed

Supported by 1 paper.

Show the supporting quotes
Here, we evaluate the log-likelihood of observing the two transit times of -i under this assumption (blue line), which is compared the distribution expected for random samplings.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

log-uniform priorUsed

Supported by 1 paper.

Show the supporting quotes
Marginalized posterior distribution (solid lines) for _L (far-left), _I (mid-left), f_L (mid-right) and f_I (far-right) using the log-uniform prior (dashed lines) and the optimistic data.
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

logistic sigmoidUsed

Supported by 1 paper.

Show the supporting quotes
We find that a logistic sigmoid (solid line in Figure ) fitted through the cross-validation results may be used to estimate R_0 _f 0 R, which provides a single-valued metric to assess the cross-validation results in what follows.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

logistic sigmoid activation functionUsed

Supported by 1 paper.

Show the supporting quotes
We repeated this exercise using a logistic sigmoid activation function instead of the rectified linear function and find nearly identical results.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Lomb-Scargle periodogram with a BIC-based definition of powerUsed

Supported by 1 paper.

Show the supporting quotes
The left panel shows the corresponding Lomb-Scargle periodogram with a BIC-based definition of power. The favored trial is plotted over the TTV data on the right panel, with the significance denoted in the corner.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

long-cadence data is resampledUsed

Supported by 1 paper.

Show the supporting quotes
Contamination factors for each quarter are accounted for using the method devised in and long-cadence data is resampled using N_resam=30 following the technique devised in .
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

low-cut, high-pass filterUsed

Supported by 1 paper.

Show the supporting quotes
This algorithm is essentially a low-cut, high-pass filter removing all long-term trends with a threshold frequency set to three times the NASA Exoplanet Archive reported transit duration.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

LS periodogram searchUsed

Supported by 1 paper.

Show the supporting quotes
Each panel includes the best-fitting sinusoidal signal as derived from a LS periodogram search.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

manually inspect an image of each and every transit light curve epochUsed

Supported by 1 paper.

Show the supporting quotes
We manually inspect an image of each and every transit light curve epoch, centred upon the time of transit minimum expected from a linear ephemeris and including 0.5 orbital periods of data either side.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

Marginalized posterior distributionUsed

Supported by 1 paper.

Show the supporting quotes
Marginalized posterior distribution (solid lines) for _L (far-left), _I (mid-left), f_L (mid-right) and f_I (far-right) using the log-uniform prior (dashed lines) and the optimistic data.
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

Marginalized posterior distributionsUsed

Supported by 1 paper.

Show the supporting quotes
Marginalized posterior distributions of M_S/M_P from model F_S for the seven KOIs investigated here, except KOI-303.01 and KOI-1472.01 for which we use models F_S,R0 and F_S,13 respectively, for reasons discussed in & .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

Markov Chain Monte Carlo (MCMC)Used

Supported by 1 paper.

Show the supporting quotes
Our regression was performed using a simple MCMC that we wrote, which used the Metropolis rule for sampling and normal proposal functions tuned by hand to give a 50% acceptance rate.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

mass_change variable in MESAUsed

Supported by 1 paper.

Show the supporting quotes
We star-lift using the mass_change variable in MESA, where this variable sets the rate of accretion per year, but allows negative values which account for mass loss.
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

maximum a posteriori transit fitUsed

Supported by 1 paper.

Show the supporting quotes
For each KOI, we export the maximum a posteriori transit fit and the corresponding vector of out-of-transit baseline fluxes (OOTs), which are found by linear minimization of the maximum a posteriori model and the data (this is actually done on the fly during every regression step, following the approach using by ).
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

maximum a-posteriori light curve fitUsed

Supported by 1 paper.

Show the supporting quotes
The first 8 panels show the Q1-9 data and the maximum a-posteriori light curve fit of a planet-only model (gray line) and a moon model (red line).
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

maximum a-posteriori planet-moon modelUsed

Supported by 1 paper.

Show the supporting quotes
For each, we overlay the maximum a-posteriori planet-moon model as conditioned upon the method marginalised light curve, and a comparison of how much better it matches the data versus the planet-only model, in a ^2 -sense.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

maximum a-posteriori solutionUsed

Supported by 1 paper.

Show the supporting quotes
We find that the maximum a-posteriori solution has a likelihood of 32.3%, highly compatible with the Big Ear data, with a broad 2\, credible interval of signal duration 72 \,secs <T<77 \,mins and mean repeat rate 0.043 \,days ^-1< <59.8 \,days ^-1 .
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

maximum likelihood planet-moon modelUsed

Supported by 1 paper.

Show the supporting quotes
Left/right column shows the first/second transit epoch, with the maximum likelihood planet-moon model overlaid in solid red.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

Maximum likelihood transit fitsUsed

Supported by 1 paper.

Show the supporting quotes
Maximum likelihood transit fits to the light curves of the 17 Earth proxies. Light curves have been folded upon the maximum likelihood ephemeris parameters and then been binned to approximately 30 points within the full transit duration.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

Maximum likelihood valueUsed

Supported by 1 paper.

Show the supporting quotes
Maximum likelihood value of 1/ _L as a function of t_L' (the earliest evidence for life).
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

MCMC algorithmUsed

Supported by 1 paper.

Show the supporting quotes
Data comes from fitting a synthetic hot-Jupiter light curve on a circular, near-grazing orbit with an MCMC algorithm.
The Transits of Extrasolar Planets with Moons (2011) DOI

MCMC inference codeUsed

Supported by 1 paper.

Show the supporting quotes
Similarly, we make the HBM posteriors available in the same repository (described in Section and presented in Figure ) as well as our MCMC inference code.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

MCMC PRF blend analysisUsed

Supported by 1 paper.

Show the supporting quotes
Summary of the results from our MCMC PRF blend analysis.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

MCMC using interpolationUsed

Supported by 1 paper.

Show the supporting quotes
Comparison of our grid of individual likelihood evaluation (left) and the emulator used for the MCMC using interpolation (right).
Could the ‘Wow’ signal have originated from a stochastic repeating beacon? (2022) DOI

median filteringUsed

Supported by 1 paper.

Show the supporting quotes
\ is applied to each transit epoch individually and the inputted data is pre-screened for any sharp, discontinuous offsets (due to say a pointing tweak) as well as a general outlier rejection phase using median filtering.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

method marginalised detrended photometryUsed

Supported by 1 paper.

Show the supporting quotes
Top: Each panel shows the method marginalised detrended photometry centred on the times of transit, with the maximum likelihood planet-moon fit overlaid in solid black.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

methods to compensate for both of these effectsIntroduced

Supported by 1 paper.

Show the supporting quotes
Chapter discusses two sources of distortion to the transit signal, namely blending (with a focus on the previously unconsidered self-blending scenario) and light curve smearing due to long integration times. I provide methods to compensate for both of these effects, thus permitting for the accurate modelling of the planetary transit light curve.
The Transits of Extrasolar Planets with Moons (2011) DOI

mock realizationsUsed

Supported by 1 paper.

Show the supporting quotes
For each transit, we generate 10,000 mock realizations where the data are drawn from perfect normal distributions at the exact same sampling observed in the data and the reported uncertainties.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

model with a single planet and variable blend factor between the two epochsUsed

Supported by 1 paper.

Show the supporting quotes
Bottom: Same as above but for a model with a single planet and variable blend factor between the two epochs.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

modified Jeffrey's priorUsed

Supported by 1 paper.

Show the supporting quotes
Unlike B12, we do not choose a fixed s value of 3 \,m\,s ^-1 but rather we fit the parameter using a modified Jeffrey's prior, as advocated by .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Modules for Experiments in Stellar Astrophysics code (MESA)Used

Supported by 1 paper.

Show the supporting quotes
We calculate mass vs. age relations using the Modules for Experiments in Stellar Astrophysics code (MESA) .
Lazarus stars: numerical investigations of stellar evolution with star-lifting as a life extension strategy (2023) DOI

Monte Carlo forward modeling of stellar atmosphere modelsUsed

Supported by 1 paper.

Show the supporting quotes
The informative priors are computed using a Monte Carlo forward modeling of stellar atmosphere models.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

Monte Carlo simulationUsed

Supported by 1 paper.

Show the supporting quotes
Panels A and B show a histogram (black) of the number of civilizations present in each snapshot of a Monte Carlo simulation with 10^6 time steps with =5.5 & =1.5 for A and =20.5 & =1.5 for B.
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

MoonPy packageIntroduced

Supported by 1 paper.

Show the supporting quotes
Much of the this analysis was carried out using tools within the MoonPy package developed by AT, available at .
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

moving median smoothing curveUsed

Supported by 1 paper.

Show the supporting quotes
We then remove all transits using the B12 ephemerides and clean the data of 3\, outliers from a moving median smoothing curve with a 20-point window (for both LC and SC data).
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

moving median trend lineUsed

Supported by 1 paper.

Show the supporting quotes
To achieve this we computed a moving median trend line for the light curve, masking the neighbor's transit (also using transit timings from , where available), and generating Gaussian-noise around the trend based on the noise profile of the light curve (see Figure ).
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

mutual eventsReviewed

Supported by 1 paper.

Show the supporting quotes
The latter, sometimes dubbed ``mutual events'' , is geometrically more probable for moons on close-in orbits.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

negative radii moonsIntroduced

Supported by 1 paper.

Show the supporting quotes
A new feature we implement in this work for the HEK project is the exploration of negative (R_S/R_P) solutions i.e. negative radii moons. Negative radii moons are of course not physically plausible and our implementation simply involves flipping the transit signal component caused by the moon in such cases. We exploit this trick as a vetting test such that solutions favoring a negative radius moon can be easily dismissed.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

negative radius exomoon vettingUsed

Supported by 1 paper.

Show the supporting quotes
In , we proposed focussing on the radius aspect of B2 alone, which we dub B2a, and testing the condition by allowing H_S to explore ``negative radius'' exomoons.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

negative-radius moonsUsed

Supported by 1 paper.

Show the supporting quotes
Another change we implement follows the strategy used recently for Kepler-22b , where we allow \ to model negative-radius moons (which we treat as inverted transits) during model S .
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

nested Keplerian bodyUsed

Supported by 1 paper.

Show the supporting quotes
At every time stamp, \ calculates the sky-projected positions of the planet and moon relative to the star, using a nested Keplerian body appropriate for a hierarchical three-body problem (see Appendix of for proof).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

nullifies the transit ingress and egressIntroduced

Supported by 1 paper.

Show the supporting quotes
An array of lasers producing a peak power of 20 \,MW for approximately 15\,minutes nullifies the transit ingress and egress.
A cloaking device for transiting planets (2016) DOI

numerical resamplingUsed

Supported by 1 paper.

Show the supporting quotes
Integration time effects are handled using numerical resampling following with N_resamp.=30 .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

numerically integratedUsed

Supported by 1 paper.

Show the supporting quotes
Next, the light curve is spline interpolated and the interpolation function is numerically integrated to compute the SNR of a single transit following Equation (15) of .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

numerically iterating on Equation ( )Used

Supported by 1 paper.

Show the supporting quotes
The solid lines are from numerically iterating on Equation ( ), whereas the dashed lines are from the simple analytic equation derived using our conjecture.
Relativistic Light Sails (2017) DOI

objective Bayesian analysis of Earth’s chronologyUsed

Supported by 1 paper.

Show the supporting quotes
Our previously proposed objective Bayesian analysis of Earth’s chronology culminated in a formula for the minimum odds ratio between the fast and slow abiogenesis scenarios (relative to Earth’s lifespan).
Strong Evidence that Abiogenesis Is a Rapid Process on Earth Analogs (2025) DOI

objective Bernoulli priorUsed

Supported by 1 paper.

Show the supporting quotes
Top: Marginalized posteriors (solid lines) for _L (far-left), _I (mid-left), f_L (mid-right) and f_I (far-right) using the objective Bernoulli prior (dashed lines), using the optimistic data.
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

odds ratio of the hypotheses that abiogenesis is a fast vs slow processUsed

Supported by 1 paper.

Show the supporting quotes
Odds ratio of the hypotheses that abiogenesis is a fast vs slow process. Each curve represents the odds ratio, computed using Equation , as a function of the biosphere's ultimate lifespan, T - although we have subtracted off the modern epoch to yield remaining time on the x -axis.
Strong Evidence that Abiogenesis Is a Rapid Process on Earth Analogs (2025) DOI

one dimensional LEBMExtended

Supported by 1 paper.

Show the supporting quotes
In this paper, we adapt the one dimensional LEBM applied to planets to function for moons with an atmosphere of similar composition to the Earth.
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

optical monochromatic laser arrayUsed

Supported by 1 paper.

Show the supporting quotes
We estimate that humanity could cloak the Earth from -like broadband surveys using an optical monochromatic laser array emitting a peak power of 30\,MW for 10 \,hours per year.
A cloaking device for transiting planets (2016) DOI

outlier rejection preconditioningUsed

Supported by 1 paper.

Show the supporting quotes
1822 KOIs for which we identify a >0 periodic TTV signal of less than 20\,cycles periodicity using the H+16 transit timing catalog (with outlier rejection preconditioning applied).
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

parameter studyUsed

Supported by 1 paper.

Show the supporting quotes
In the second investigation, we carry out an extensive parameter study, running the model many times to study the habitability of the exomoon in the 4-dimensional space composed of the planet semi-major axis and eccentricity, and the moon semi-major axis and eccentricity.
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

Periodogram of the TTVsUsed

Supported by 1 paper.

Show the supporting quotes
Bottom: Periodogram of the TTVs and the FAP =50 % threshold, which highlights the absence of any significant TTVs here.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

periodograms of the TTVs & TDVsUsed

Supported by 1 paper.

Show the supporting quotes
Bottom panels show the periodograms of the TTVs & TDVs (which use the entire available data set).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

phase folded lightcurveUsed

Supported by 1 paper.

Show the supporting quotes
A phase folded lightcurve and accompanying dialog box.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

phase shiftUsed

Supported by 1 paper.

Show the supporting quotes
By detecting both signals and observing this phase shift, the detection would be unambiguous against all other phenomenon, since presently no other effect is predicted to induce such a phase shift.
The Transits of Extrasolar Planets with Moons (2011) DOI

phase-folding upon the candidate TTV periodUsed

Supported by 1 paper.

Show the supporting quotes
After phase-folding upon the candidate TTV period, the TTVs are disproportionally occuring in the second-half of the diagram.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

phase-foldsUsed

Supported by 1 paper.

Show the supporting quotes
It is also worth keeping in mind that moon transits are not always readily apparent in phase-folded transits, as the moon transit will appear in a different location each epoch; we present phase-folds here for want of a cleaner way to display so many individual transits.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

Photodynamic Look-Up Tables (LUTs)Used

Supported by 1 paper.

Show the supporting quotes
Photodynamic Look-Up Tables (LUTs) As a result of the myriad of complicated effects influencing the final model yet the relative low-dimensionality of the model itself, we elected to build a grid of pre-computed models which accounted for all of the effects described above.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

photodynamical modelingUsed

Supported by 1 paper.

Show the supporting quotes
Using photodynamical modeling and Bayesian multimodal nested sampling, we find no compelling evidence for an exomoon in these eight systems.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

photodynamical planet-moon codeUsed

Supported by 1 paper.

Show the supporting quotes
For this reason, we model the six transits of Kepler-90g using the photodynamical planet-moon code \ plus six freely fitted transit times, which are likely due to planet-planet interactions.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

photodynamical search for an exomoonUsed

Supported by 1 paper.

Show the supporting quotes
In this work, we conduct a photodynamical search for an exomoon around \ leveraging the transits, radial velocities and asteroseismology plus several new tools developed by the HEK project to improve exomoon searches.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

PIKAIAUsed

Supported by 1 paper.

Show the supporting quotes
The fitting code finds the best-fit to the light curve by utilising a genetic algorithmAvailable from http://whitedwarf.org/parallel/, PIKAIA , to get close to a minimum in ^2 .
The Transits of Extrasolar Planets with Moons (2011) DOI

planet fit on each individual transitUsed

Supported by 1 paper.

Show the supporting quotes
Finally, we perform a planet fit on each individual transit to derive TTVs and TDVs (model I ), which is useful in comparing our moon model against perturbing planet models later on.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

planet-with-moon light curve modelling routineUsed

Supported by 1 paper.

Show the supporting quotes
Selected targets are fitted using a multimodal nested sampling algorithm coupled with a planet-with-moon light curve modelling routine.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

Poisson processUsed

Supported by 1 paper.

Show the supporting quotes
We follow the prescription of in adopting the simple yet reasonable assumption that abiogenesis events occur at a uniform rate over time, thus following a Poisson process.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

polynomial detrendingsUsed

Supported by 1 paper.

Show the supporting quotes
We confirmed that this signal is coherent in both the PDC and Simple Aperture Photometry (SAP) data products, using both \ and polynomial detrendings of each.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

population-weighted main-sequence lifetimesUsed

Supported by 1 paper.

Show the supporting quotes
two orders-of-magnitude below the expected value by comparing the population-weighted main-sequence lifetimes.
Formulation and resolutions of the red sky paradox (2021) DOI

Posterior probability distributionUsed

Supported by 1 paper.

Show the supporting quotes
Posterior probability distribution of (solid) when conditioned upon the timing of Earth's evolutionary record.
Formulation and resolutions of the red sky paradox (2021) DOI

Pre-search Data Conditioning (PDC) time seriesUsed

Supported by 1 paper.

Show the supporting quotes
The Pre-search Data Conditioning (PDC) time series was downloaded and detrended after first excluding all transits, outliers and manually removing discontinuities (e.g. ramps), in the same way described in our previous papers (e.g. see ).
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

Precision-recall plotUsed

Supported by 1 paper.

Show the supporting quotes
Precision-recall plot of the CNN ensemble on (simulated) validation data, indicating that maximum precision is acquired when the ensemble is also in total agreement.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

principal components analysis (PCA)Used

Supported by 1 paper.

Show the supporting quotes
To this end, we perform a principal components analysis (PCA) on the joint prior probability of u_1 - u_2 and re-parameterize the terms into the orthogonal components w_1 and w_2 following the suggestion of :
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

principle of ensemble equivalenceIntroduced

Supported by 1 paper.

Show the supporting quotes
We show how the sail's motion in response to an ensemble of incident photons is equivalent to that of a single photon of energy equal to that of the ensemble. We use this ``principle of ensemble equivalence'' for both perfect and imperfect mirrors, enabling a simple analytic prediction of the sail's velocity curve.
Relativistic Light Sails (2017) DOI

probabilistic mass-radius relationUsed

Supported by 1 paper.

Show the supporting quotes
Training data used to infer our probabilistic mass-radius relation for late-type stars. Black points are 171 late-type stars with empirically determined masses and radii taken from previous work .
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

q_1 - q_2 parameter setUsed

Supported by 1 paper.

Show the supporting quotes
The only change from the priors, is that we fit the quadratic limb darkening coefficients using the q_1 - q_2 parameter set suggested in , which provides more efficient parameter exploration.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

quadratic limb darkening formulationUsed

Supported by 1 paper.

Show the supporting quotes
In this work, we use the quadratic limb darkening formulation described in and account for long-cadence smearing using the re-sampling treatment described in ( N_resam=30 ).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

R package entropyUsed

Supported by 1 paper.

Show the supporting quotes
For this work, we use an R package entropy with all the KLD calculations.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

randomly selectedUsed

Supported by 1 paper.

Show the supporting quotes
We then repeat this same process except the follow-up set is randomly selected, and the number of success is saved as S_R . The ratio of these values, R , provides an estimate of how much better our ANN performs versus random picks, where R=S_ANN/S_R.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

rank-ordering the objects by their class probabilitiesUsed

Supported by 1 paper.

Show the supporting quotes
We then define some follow-up fraction, f , and after rank-ordering the objects by their class probabilities, pick only the top f -quantile.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

Rayleigh distributionUsed

Supported by 1 paper.

Show the supporting quotes
Here, both the injected and inferred eccentricity distributions are the Rayleigh distribution.
Near-circular orbits for planets with Earth-like sizes and instellations around M and K dwarf stars (2025) DOI

re-sampling treatmentUsed

Supported by 1 paper.

Show the supporting quotes
In this work, we use the quadratic limb darkening formulation described in and account for long-cadence smearing using the re-sampling treatment described in ( N_resam=30 ).
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). V. A SURVEY OF 41 PLANETARY CANDIDATES FOR EXOMOONS (2015) DOI

re-scale the time axis into distance from the planetUsed

Supported by 1 paper.

Show the supporting quotes
Instead, in this work, we argue a better approach is to re-scale the time axis into distance from the planet, in units of planetary radii.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

regressed quadratic functionsUsed

Supported by 1 paper.

Show the supporting quotes
We largely see this behaviour but the contours get very close together so to ensure this behaviour against numerical error we regressed quadratic functions in each b -slice and use these predictions to slightly refine the contours.
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

resamplingUsed

Supported by 1 paper.

Show the supporting quotes
However, here we resample to a full short-cadence resolution using N_resam = 30 .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

RMS vs bin testUsed

Supported by 1 paper.

Show the supporting quotes
Bottom: RMS vs bin test of the data, showing the four relevant regions with separate black lines and the expected range for Gaussian statistics in blue.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

Savage-Dickey ratioUsed

Supported by 1 paper.

Show the supporting quotes
We next evaluate the Savage-Dickey ratio between the posterior and the prior to estimate the Bayes factor, Z_circ/Z_ecc .
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

search for statistically significant excess varianceUsed

Supported by 1 paper.

Show the supporting quotes
The only available method is to therefore search for statistically significant excess variance and then use the ^2 -distribution to calculate the confidence of signal detection.
The Transits of Extrasolar Planets with Moons (2011) DOI

signal injection and blind retrievalUsed

Supported by 1 paper.

Show the supporting quotes
By signal injection and blind retrieval, we demonstrate that an Earth-like moon is easily detected for this planet even when the time-correlated noise of the data set is taken into account.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

signal to noise ratio (SNR) of the planetary transit signal in each pixelUsed

Supported by 1 paper.

Show the supporting quotes
Middle: Same as the left, except we show the signal to noise ratio (SNR) of the planetary transit signal in each pixel.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

significance and robustness checksUsed

Supported by 1 paper.

Show the supporting quotes
Here, we search through published \ TTV data for such signals, applying a battery of significance and robustness checks, plus independent light curve analyses for candidate signals.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

simple analytic equationIntroduced

Supported by 1 paper.

Show the supporting quotes
The solid lines are from numerically iterating on Equation ( ), whereas the dashed lines are from the simple analytic equation derived using our conjecture.
Relativistic Light Sails (2017) DOI

Simple Aperture Photometry (SAP) data productsUsed

Supported by 1 paper.

Show the supporting quotes
We confirmed that this signal is coherent in both the PDC and Simple Aperture Photometry (SAP) data products, using both \ and polynomial detrendings of each.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

simple clipping techniquesUsed

Supported by 1 paper.

Show the supporting quotes
Short-term variations such as flaring, pointing tweaks and safe-mode recoveries (charge trapping) are removed manually by simple clipping techniques.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

simple majority voteUsed

Supported by 1 paper.

Show the supporting quotes
A total of 65 KOIs (3.5%) were classified as having more than 50% of their transits containing moons by a simple majority vote.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

truncation to the PDFUsed

Supported by 1 paper.

Show the supporting quotes
The selection bias introduced by can be formally encoded into our model by applying a truncation to the PDF for , such that Pr( ; ) &= e^- 1 - e^- .
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

simple model to predict the sail's temperature and diffraction beam lossesUsed

Supported by 1 paper.

Show the supporting quotes
Additionally, we use a simple model to predict the sail's temperature and diffraction beam losses during the laser firing period, allowing us to estimate that for firing times of a few minutes and operating temperatures below 300^ C ( 573 K), Starshot will require a sail of which absorbs less than 1 in 260,000 photons.
Relativistic Light Sails (2017) DOI

Simulated difference imagesUsed

Supported by 1 paper.

Show the supporting quotes
Simulated difference images for the various stars compared with the observed difference image.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

simulated transitsReviewed

Supported by 1 paper.

Show the supporting quotes
(Left) Six simulated transits using \ of a HZ Neptune around an M2 star with an Earth-like moon on a wide orbit (90% of the Hill radius).
In Search of Exomoons (2014) DOI

simulating posterior distributions and resulting entropic information gainsUsed

Supported by 1 paper.

Show the supporting quotes
By simulating posterior distributions and resulting entropic information gains, we compare three experimental pressures on : 1) evidence for an earlier start to life; t_obs ; 2) constraints on spontaneous abiogenesis from the lab; and 3) an exoplanet survey for biosignatures.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

Single-body Asterodensity ProfilingUsed

Supported by 1 paper.

Show the supporting quotes
This trick was first noted in and was dubbed ``Single-body Asterodensity Profiling'' by the authors, or SAP. We are therefore able to provide SAP-derived eccentricity constraints in the light curve fits provided in this work.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

single-layer U_1=4 ANNUsed

Supported by 1 paper.

Show the supporting quotes
These results imply that a single-layer U_1=4 ANN is sufficient to capture the predictive power of the selected features.
Transit clairvoyance: enhancingTESSfollow-up using artificial neural networks (2016) DOI

smooth kernel density estimator (KDE) with a Gaussian kernelUsed

Supported by 1 paper.

Show the supporting quotes
Left: We extract a random draw from the eccentricity posterior distribution of each planet and apply a smooth kernel density estimator (KDE) to the sample with a Gaussian kernel.
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

smoothed functional fitUsed

Supported by 1 paper.

Show the supporting quotes
Empirical sensitivity of the HEK project to the satellite planet mass ratio (red line), based on 17 published results. HEK is able to detect Earth-Moon like mass ratios in 14 % of cases. Black line shows a smoothed functional fit.
In Search of Exomoons (2014) DOI

SNR equationExtended

Supported by 1 paper.

Show the supporting quotes
These three constraints are the same as those in except that we slightly modify the SNR equation to account for transit duration more appropriately using SNR_i &= (R_ /R_ )^2 CDPP_6 T_146\,hours, & = SNR_i N_transits,
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

solar blackbodyUsed

Supported by 1 paper.

Show the supporting quotes
Bottom: Power requirement P_t ( ) in W, assuming a solar blackbody with T_eff = 5777\,K and chromatic lasers spaced dynamically according to Equation ( ) and setting R=1000 .
A cloaking device for transiting planets (2016) DOI

spline interpolatedUsed

Supported by 1 paper.

Show the supporting quotes
Next, the light curve is spline interpolated and the interpolation function is numerically integrated to compute the SNR of a single transit following Equation (15) of .
The giant nature of WD 1856 b implies that transiting rocky planets are rare around white dwarfs (2023) DOI

split up our light curves in segments of 10 epochsUsed

Supported by 1 paper.

Show the supporting quotes
Instead, we split up our light curves in segments of 10 epochs each, providing a manageable number of free parameters for each.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

stack the phase-folded transitsUsed

Supported by 1 paper.

Show the supporting quotes
In this work, we stack the phase-folded transits of 284 viable moon hosting Kepler planetary candidates, in order to search for satellites.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

stochastic formulationExtended

Supported by 1 paper.

Show the supporting quotes
Here, I build upon earlier work and present a stochastic formulation.
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

sum of harmonic cosinesUsed

Supported by 1 paper.

Show the supporting quotes
Periodic functions described by a sum of harmonic cosines are explored from twice the data baseline down to twice the transit duration, with a cap of 30 harmonics (beyond which we tend to encounter numerical instabilities).
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

synthetic light curvesUsed

Supported by 1 paper.

Show the supporting quotes
Having modified W to W' to account for these additional noise sources, I proceed to test this modified formulation through generation of synthetic light curves.
The Transits of Extrasolar Planets with Moons (2011) DOI

synthetic stellar signalUsed

Supported by 1 paper.

Show the supporting quotes
For the stellar noise, 1000 sinusoidal waveforms are generated with varying periods randomly selected between one minute up to 24 hours. These waveforms are then coadded to give a single synthetic stellar signal designed to mimic the non-stochastic nature of real stellar variability.
The Transits of Extrasolar Planets with Moons (2011) DOI

synthetic, planet-only and planet+moon single transitsUsed

Supported by 1 paper.

Show the supporting quotes
Our training set consists of 27,000 examples of synthetic, planet-only and planet+moon single transits, injected into \ light curves.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

Target Selection Automatic (TSA) algorithmUsed

Supported by 1 paper.

Show the supporting quotes
In this work, Target Selection (TS) is conducted using the Target Selection Automatic (TSA) algorithm described in and updated in .
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

target selection prioritization (TSP) filteringUsed

Supported by 1 paper.

Show the supporting quotes
These seven candidates were identified through our automatic target selection (TSA) algorithm and target selection prioritization (TSP) filtering, whereby we excluded systems exhibiting significant time-correlated noise and focussed on those with a single transiting planet candidate of radius less than 6 \,R_ .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

test for uniform phase coverageUsed

Supported by 1 paper.

Show the supporting quotes
Top: an example of a TTV signal that fails our test for uniform phase coverage.
A search for transit timing variations within the exomoon corridor using Kepler data (2022) DOI

the nested two-body modelUsed

Supported by 1 paper.

Show the supporting quotes
In this work, I will adopt what I label as ``the nested two-body'' model. In this framework, one has the planet-moon forming a close-binary which orbits the star at a much greater distance.
The Transits of Extrasolar Planets with Moons (2011) DOI

time based measurementUsed

Supported by 1 paper.

Show the supporting quotes
A time based measurement. Here, the duration of a candidate signal is being measured.
LcTools: A Windows-Based Software System for Finding and Recording Signals in Lightcurves from NASA Space Missions (2019) DOI

time-averaged valueUsed

Supported by 1 paper.

Show the supporting quotes
Each simulation is allowed to evolve until it reaches a steady or quasi-steady state, and the final ten years of climate data are used to produce a time-averaged value of (t) , , and the sample standard deviation, _ .
Dynamical effects on the habitable zone for Earth-like exomoons (2013) DOI

transit centroidIntroduced

Supported by 1 paper.

Show the supporting quotes
In order to investigate this, we considered a new type of centroid we dub the ``transit centroid''. Whereas a conventional centroid calculates the mean pixel position weighted by the flux of each pixel, the transit centroid instead weights by the signal-to-noise ratio (SNR) of the transit depth on each pixel.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

transit duration variations (TDV)Introduced

Supported by 1 paper.

Show the supporting quotes
The previously known TTV effect is updated with a new model and the associated critical problems are outlined. I then predict a new effect, TDV, which solves these problems, making exomoon detection viable.
The Transits of Extrasolar Planets with Moons (2011) DOI

Transit impact parameter induced transit duration variations (TDV-TIP)Reviewed

Supported by 1 paper.

Show the supporting quotes
Transit impact parameter induced transit duration variations (TDV-TIP) Later, showed that a second transit duration variation (TDV) effect exists, dubbed TDV-TIP, standing for transit-impact-parameter induced TDV.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

Transit timing and duration variations (TTV and TDV)Reviewed

Supported by 1 paper.

Show the supporting quotes
Transit timing and duration variations (TTV and TDV) exhibit complementary sensitivities with the period-ratio.
In Search of Exomoons (2014) DOI

transit timing variationsUsed

Supported by 1 paper.

Show the supporting quotes
For KOI-314, we are able to confirm the planetary nature of two out of the three known transiting candidates using transit timing variations.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). IV. A SEARCH FOR MOONS AROUND EIGHT M DWARFS (2014) DOI

transit timing variations (TTV)Extended

Supported by 1 paper.

Show the supporting quotes
Before I provide this new method, I will first provide an updated model for the transit timing variations (TTV) of an exomoon, including the effects of orbital eccentricity and inclination of both the planet and moon plus the mutual longitude of the ascending node.
The Transits of Extrasolar Planets with Moons (2011) DOI

Transit timing variations (TTV)Reviewed

Supported by 1 paper.

Show the supporting quotes
Transit timing variations (TTV) The first conceived effect was transit timing variations (TTV), by , which is conceptually analogous to the astrometric technique of finding planets around stars.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

transit timing variations (TTVs)Used

Supported by 1 paper.

Show the supporting quotes
We use colour (see legend) to depict the inferred absence/presence of transit timing variations (TTVs).
An exomoon survey of 70 cool giant exoplanets and the new candidate Kepler-1708 b-i (2022) DOI

transiting moonsReviewed

Supported by 1 paper.

Show the supporting quotes
Further, in an analogous manner to how transiting planets may yield the mean stellar density, _* , , transiting moons allow one to measure the mean planetary density directly, _P , as first shown in .
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

Trend Filtering Algorithm (TFA)Used

Supported by 1 paper.

Show the supporting quotes
As a further check, we tried detrending the Q6 photometry using the Trend Filtering Algorithm (TFA) described in , which searches for common mode behaviors on the same CCD module.
THE POSSIBLE MOON OF KEPLER-90g IS A FALSE POSITIVE (2015) DOI

triangular samplingUsed

Supported by 1 paper.

Show the supporting quotes
Employing triangular sampling is approximately twice as efficient than the uninformative priors used previously in .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

TTV waveform and RMS amplitudeExtended

Supported by 1 paper.

Show the supporting quotes
In the limit of circular and co-aligned orbits, as was considered by , _TTV and thus the RMS amplitude agrees with that found by the original paper. However, the advantages of this much more general expression for the TTV waveform and RMS amplitude are manifold.
The Transits of Extrasolar Planets with Moons (2011) DOI

TTVs and TDVsUsed

Supported by 1 paper.

Show the supporting quotes
TTVs (black circles) and TDVs (gray squares) from model fit V_V of the seven KOIs investigated here. TTVs computed relative to the maximum a-posteriori ephemeris derived by model F_P . TDVs computed relative to the median duration from model F_P .
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

two-pass harmonic detrendingUsed

Supported by 1 paper.

Show the supporting quotes
Our data processing includes two-pass harmonic detrending, transit timing variations, model selection and careful data quality vetting to produce a grand light curve with a r.m.s. of 5.1\,ppm.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

uniform distributionUsed

Supported by 1 paper.

Show the supporting quotes
Thus the least informative prior is a uniform distribution given by Pr( ) = U[1.5,4.5], and this is the distribution adopted in what follows.
On the Rate of Abiogenesis from a Bayesian Informatics Perspective (2018) DOI

uniform prior in fUsed

Supported by 1 paper.

Show the supporting quotes
Lower: same as top panel but using the uniform prior in f for comparison.
An objective Bayesian analysis of life’s early start and our late arrival (2020) DOI

uniform rate (Poisson) stochastic processUsed

Supported by 1 paper.

Show the supporting quotes
The birth of civilizations within the galaxy is modeled as following a uniform rate (Poisson) stochastic process, with a mean rate of .
A Stationary Drake Equation Distribution as a Balance of Birth-death Processes (2021) DOI

Velocity induced transit duration variations (TDV-V)Reviewed

Supported by 1 paper.

Show the supporting quotes
Velocity induced transit duration variations (TDV-V) TDV-V was conceived of a decade after TTV by . showed that the same motion responsible for changes in position causing TTV should also cause changes in velocity.
THE HUNT FOR EXOMOONS WITHKEPLER(HEK). I. DESCRIPTION OF A NEW OBSERVATIONAL PROJECT (2012) DOI

Visual inspection of the feature mapsUsed

Supported by 1 paper.

Show the supporting quotes
Visual inspection of the feature maps for these KOIs (see section 2.4.3) reveals that the moon classifications for these systems, like the simulated light curves, are due to the planet (candidate) transit morphologies.
Identifying potential exomoon signals with convolutional neural networks (2021) DOI

Von Mises distributionUsed

Supported by 1 paper.

Show the supporting quotes
The inclination of the moons are randomly drawn from a Von Mises distribution with =42637 , which we found to maximize the likelihood of a Von Mises distribution conditioned on the real Galilean moons.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

weighted linear minimizationUsed

Supported by 1 paper.

Show the supporting quotes
This means that we can employ weighted linear minimization, which is not only computationally quicker than non-linear methods, but also guaranteed to reach the global minimum. In our regression, the data are weighted by the inverse of their reported standard photometric errors.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). II. ANALYSIS OF SEVEN VIABLE SATELLITE-HOSTING PLANET CANDIDATES (2013) DOI

weighted linear minimizationExtended

Supported by 1 paper.

Show the supporting quotes
In this work, we introduce a new refinement to the HEK fitting strategy which simply treats the baseline parameters ( OOT_j ) as a set of nuisance parameters. Rather than marginalizing over these terms, every realization of the transit model simply optimizes the baseline parameters to the data. This is easily achieved using a weighted linear minimization which is both extremely quick and guaranteed to always find the global minimum.
THE HUNT FOR EXOMOONS WITH KEPLER (HEK). III. THE FIRST SEARCH FOR AN EXOMOON AROUND A HABITABLE-ZONE PLANET (2013) DOI

weighted linear regressionUsed

Supported by 1 paper.

Show the supporting quotes
Instead, we use weighted linear regression to find the maximum likelihood linear ephemeris through the posterior transit times and then inspect the residuals for evidence of TTVs.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI

wget scriptUsed

Supported by 1 paper.

Show the supporting quotes
Using a wget script, any and all of the target's LC data is then downloaded from MAST (from archive.stsci.edu/pub/kepler/lightcurves/) and saved in the appropriate directory.
HEK. VI. On the Dearth of Galilean Analogs in Kepler, and the Exomoon Candidate Kepler-1625b I (2017) DOI