Pith. sign in

REVIEW 4 major objections 6 minor 106 references

High-Pass Filtering and Gaussian Process Regularization: Stellar Activity Characterization Techniques Applied to the 55 Cancri Planetary System

T0 review · 4 major / 6 minor · reviewed 2026-08-04 · deepseek-v4-flash

Pith's one-line read The paper claims that a four-planet model of 55 Cancri, with stellar activity handled by a curvature-penalized Gaussian process, matches the standard five-planet model, so planet d's existence is not required by the radial velocities alone.

desk verdict A useful GP regularization idea with an honest application, but the 4-planet/5-planet comparison rests on a hand-set penalty parameter that differs between models. read the letter →

arxiv 2509.08076 v1 pith:Z63HQ2M3 submitted 2025-09-09 astro-ph.EP astro-ph.IM

classification astro-ph.EPastro-ph.IM
keywords 55CancriradialvelocityexoplanetsstellaractivityGaussianprocessfiltermodelcomparisonplanetd
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that the long-period radial-velocity signal in 55 Cancri, previously split between a fifth planet (d) and a periodically modeled magnetic cycle, can be explained by a quasiperiodic stellar activity cycle alone. To test this, the authors introduce two data-analysis tools: a Gaussian high-pass filter that removes long-term trends without needing a model for them, and a curvature penalty added to the Gaussian-process likelihood that stops the activity model from overfitting. Fitting the archival radial velocities, a four-planet model (b, c, e, f) plus a regularized GP for activity matches the standard five-planet model in goodness of fit, and outperforms it by some metrics. The paper concludes that the existence of planet d cannot be established from radial velocities alone, and that independent observations are required.

What carries the argument

The carrying object is the curvature-penalized Gaussian process: a quasiperiodic GP kernel (squared-exponential times exponential-sine-squared, from Rajpaul et al. 2015) whose negative log-likelihood is augmented by a term proportional to the sum of squared second differences of the predicted activity signal. This penalty prevents the GP from interpolating short-term noise and confines it to dynamo-timescale variation. A second tool, the Gaussian high-pass filter, computes a weighted moving average trend on unevenly spaced timestamps and subtracts it, so short-period planet signals can be found in periodograms without committing to a model of the long-term variability.

What would settle it

Re-run the 4pGP versus 5pGP comparison with the curvature penalty alpha fixed by an objective rule, such as cross-validation, applied identically to both models; the paper set alpha by visual inspection and used different values for the two models, so a robust winner under a single principled alpha would settle the model-comparison claim. Independently, Gaia astrometry can decide planet d: the 5pGP fit predicts an astrometric wobble of about 1.4 mas on a ~4600-day orbit, well above Gaia's precision, so a null detection at that amplitude would support the four-planet model.

Watch

Extended reading notes

Core claim

The central claim is that modeling stellar magnetic activity with a quasiperiodic Gaussian process, regularized by a curvature penalty, removes the need for the disputed fifth planet in 55 Cancri. In the paper's comparison, the 4pGP model—Keplerian orbits for planets b, c, e, and f plus a GP for the activity cycle—achieves reduced chi-squared 0.75 and residual RMS 5.49 m/s, at least as good as the 5pGP model (0.80, 5.82 m/s) and comparable to the previously published 5-planet model (0.81, 5.38 m/s). The GP activity cycle in the four-planet model has a period of about 5000 days and amplitude ~12 m/s, consistent with earlier estimates of the magnetic cycle amplitude, while the five-planet mode

Load-bearing premise

The load-bearing premise is that the curvature penalty alpha, chosen by visual inspection and set to different values for the two models, fairly represents how flexible the activity Gaussian process should be; if the alpha values are arbitrary, the comparison between four- and five-planet models is not decisive.

Editorial extensions

If this is right

  • If the 4pGP model is accepted, 55 Cnc has four confirmed planets and planet d becomes an unconfirmed candidate whose parameters are not well determined.
  • The activity-cycle period of ~5000 days found by the GP is not consistent with the 3822-day strictly periodic activity model, implying that treating magnetic cycles as Keplerians can bias long-period planet parameters.
  • A reanalysis of other systems with long-period planets and active stars should check whether a regularized GP can absorb the supposed outer planet signal before claiming a detection.
  • Future Gaia astrometry can decide the question: the ~1.4 mas wobble predicted for planet d is measurable if it exists.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The curvature-penalty choice is the hinge of the result; a principled, automatic selection of alpha (e.g., cross-validation) applied equally to both models would make the 4-vs-5 planet comparison more decisive. The paper itself relies on visual inspection and uses different alpha values for the two models.
  • The same overfitting guard could be applied to other quasiperiodic stellar signals in RV surveys, not just long magnetic cycles, wherever a GP risks absorbing a planet.
  • If astrometry does confirm planet d, the comparison would not be wasted: the exercise would still show that RV data alone cannot separate a 5000-day activity cycle from a ~4600-day planet, and that combined RV+astrometry fitting is needed.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This paper develops two tools for RV planet searches: high-pass detrending with a Gaussian filter (GF) and a quasiperiodic Gaussian-process activity model fitted with a curvature-penalized objective (Eq. 5). The methods are applied to the 55 Cnc RV dataset of Bourrier et al. (2018). After GF detrending and periodogram searches, the authors fit two global models: 4pGP (planets b, c, e, f plus a GP for activity) and 5pGP (b, c, e, f, d plus a GP). They report similar goodness-of-fit metrics for the two models and conclude that planet d is not required by the RVs and cannot be established from RVs alone. The paper includes a public Jupyter notebook and a Zenodo archive, which is a genuine strength.

Significance. The strongest contribution is methodological: GF detrending is simple, transparent, and useful when no physical model of the long-term trend is available, and the curvature-penalized GP objective is a sensible response to the known problem of GP overfitting. The authors are also unusually candid about the heuristic nature of their choices. If the conclusion about planet d were robust, it would have substantial consequences for the 55 Cnc system. However, the central model comparison is currently not controlled: the two models use different regularization strengths, the quoted objective values are not comparable likelihoods, and the jitter treatment is post-hoc. The planet-d conclusion is therefore an interesting hypothesis rather than an established result.

major comments (4)
  1. [Eq. (5), Table 3, Sec. 6.2] The 4pGP and 5pGP models are optimized with different curvature penalties: α = 2×10^8 for 4pGP and α = 3×10^9 day^4 m^-2 s^2 for 5pGP. Since L(θ) in Eq. (5) includes the penalty term, the reported values −6123 and −6370 are not comparable log-likelihoods; the 5pGP GP is penalized 15 times more strongly, artificially limiting its ability to absorb long-period variability. The statement in Sec. 6.2 that the models have 'comparable L(θ)' is therefore unsupported. A controlled comparison is needed: for example, the same α for both models, the pure unpenalized log-likelihood, or cross-validated predictive scores.
  2. [Sec. 4.2, Sec. 6.2] The values of α are chosen by visual inspection, separately for each model, as the minimum penalties needed to prevent short-term GP interpolation. No sensitivity analysis is reported. Because the central claim is that a GP can replace planet d, the dependence of that claim on α must be quantified. The paper should show how the best-fit parameters and goodness-of-fit metrics vary over a broad range of α for both models, or adopt an algorithmic selection procedure (e.g., cross-validation) applied identically to 4pGP and 5pGP.
  3. [Table 3, Fig. 5] The two models place the long-period RV power in very different components: the 4pGP GP has K = 12.2 ± 6.7 m/s, while 5pGP gives planet d K = 38.6 ± 3.5 m/s and a GP amplitude of only 4.3 ± 3.0 m/s. A quasiperiodic GP with marginal standard deviation ~12 m/s will only rarely produce a coherent 38 m/s signal, so the comparison is not simply 'the GP absorbs planet d'; the two models are fitting different decompositions of the long-term variation. An injection-recovery test or posterior predictive check would clarify whether the 4pGP GP can actually generate the long-period structure that 5pGP attributes to planet d.
  4. [Appendix A, Tables 1 and 3] Jitter is estimated post-hoc from the residuals of each fit and then used to compute reduced χ² and the Anderson-Darling statistics. The resulting difference in reduced χ² (0.75 vs 0.80) is small and is quoted without uncertainty, and values below unity indicate overfitting or overestimated jitter. The claim that 4pGP is 'statistically superior' to 5pGP is too strong. A joint fit with jitter parameters (or an appropriate marginalization/bootstrap) is needed before this comparison can support the paper's conclusion.
minor comments (6)
  1. [Sec. 5.1, Sec. 7] The absence of a planet-d peak in the periodograms is not informative because the GF with σ = 500 days explicitly removes variability on timescales ≳1000 days, and the paper itself notes this in Sec. 5.1. The later statement in Sec. 7 that there is 'no evidence for planet d in the frequency domain' is misleading without additional analysis of the unfiltered residuals.
  2. [Sec. 6.2] The expression for reduced χ² appears to be missing squares: it is written as [Σ (Δy_n − τ_n)/σ_y,n]/(N−γ). The standard definition is (1/(N−γ)) Σ [(Δy_n − τ_n)/σ_y,n]^2. Please correct the typo.
  3. [Sec. 6.1, Sec. 7] Minor language issues: 'modest affect' should be 'modest effect', and 'principle concern' should be 'principal concern'.
  4. [Software/References] The Zenodo DOI in the text is 10.5281/zenodo.14571274, but the reference list gives 10.5281/zenodo.14571275. Please make these consistent.
  5. [Sec. 6.2] The two-sample Anderson-Darling test statistic is reported as −1.20 with p = 0.99998 identically for both models. Negative statistic values and identical values for two models with different residuals need explanation; this may be a reporting artifact.
  6. [Eq. (5)] The curvature penalty uses central differences at sampling points m=0,...,M−1; please specify how boundary points are handled when computing τ_{m+1}−2τ_m+τ_{m−1} at the edges of the grid.

Circularity Check

1 steps flagged · score 3.0 of 10

Partial circularity in jitter-based goodness-of-fit; central planet-d claim is a valid degeneracy test but rests on subjective regularization.

  1. fitted input called prediction [Appendix A (jitter estimation), applied in §6.2 and Table 3]
    "If jitter were not a consideration, σ_y,n would fully capture the uncertainty in y_n, so we would expect 1/(N−γ) Σ ((Δy_n−τ_n)/σ_y,n)^2 ≈ 1... Fitting a Gaussian distribution N(0,W) to the δy distribution for each instrument yields a weighting factor W that quantifies the degree to which σ_y underestimates the uncertainty in y_n."

    The jitter σ_J is estimated from each model's own residuals by forcing the normalized residual variance to be approximately 1 (Equation A1). The reduced χ² values in Table 3 (0.75 for 4pGP, 0.80 for 5pGP) are then presented as goodness-of-fit evidence favoring 4pGP, but they are near unity by construction for both models. Thus the reduced-χ² comparison is not an independent test; it is a fitted quantity being used as a model-quality prediction. The residual RMS values (5.49 vs 5.82 m/s) are not circular, but the statistical superiority claim leans on a metric that is partly self-constructed.

full rationale

The paper's central claim—that a 4-planet model with a GP activity component performs as well as the 5-planet model—is a model-comparison result, not a derivation from a fitted parameter. The analysis uses public RV data, external codes (george, kepmodel, DACE), and explicitly compares against Bourrier et al. (2018). The quasiperiodic kernel is taken from Rajpaul et al. (2015) and the curvature penalty from Green & Silverman (1994), both independent of the authors. The conclusion that planet d cannot be established from RVs alone is not forced by a self-citation chain; it is a genuine degeneracy statement. However, there is partial circularity in the goodness-of-fit assessment: jitter is estimated from the residuals of each model (Appendix A) in a way that drives reduced χ² toward 1, so the reduced-χ² comparison is not fully independent. Additionally, the curvature penalty α is chosen separately for 4pGP and 5pGP by visual inspection (§4.2), making the L(θ) values in Table 3 non-comparable across models. These are methodological weaknesses and a modest statistical circularity, but the core claim about planet d is not equivalent to the model inputs. Score 3 reflects partial, non-central circularity.

Assumptions & free parameters 8 free parameters · 4 assumptions · 0 invented entities

The central claim depends on the hand-chosen alpha values and the assumed GP kernel shape. No new physical entities are introduced.

free parameters (8)
  • alpha (4pGP) = 2e8 day4 m-2 s2
    Chosen by visual inspection as the minimum curvature penalty to prevent GP interpolation of short-term fluctuations (Sec 4.2).
  • alpha (5pGP) = 3e9 day4 m-2 s2
    Chosen similarly for the 5-planet model; differs from 4pGP alpha, confounding model comparison.
  • GP amplitude K (4pGP) = 12.2 +/- 6.7 m/s
    Fitted GP hyperparameter in Table 3.
  • GP period P (4pGP) = 4980 +/- 930 days
    Fitted GP hyperparameter in Table 3.
  • lambda_e (4pGP) = 5015 +/- 3950 days
    Fitted GP hyperparameter in Table 3.
  • lambda_p (4pGP) = 0.24 +/- 0.062
    Fitted GP hyperparameter in Table 3.
  • Gaussian filter width sigma = 500 days
    Chosen by hand; authors state the trend changes marginally for sigma in [400,1000] days (Sec 3.2).
  • Instrument jitter estimates = Various, Table 1
    Estimated post-hoc using CDF fitting and an analytical approximation (Appendix A).
assumptions (4)
  • domain assumption The quasiperiodic GP kernel of Rajpaul et al. (2015) is an appropriate model for stellar magnetic activity cycles.
    Invoked in Sec 4.1 via Eq 3; central to the activity model.
  • domain assumption The long-term radial velocity variability is quasiperiodic and can be represented by a single GP without an additional Keplerian.
    Underpins the 4pGP model; stated in Sec 5.2.
  • domain assumption The archival RV dataset and error bars from Bourrier et al. (2018) are reliable.
    The entire analysis uses this public dataset (Sec 2).
  • standard math Planets obey standard Keplerian orbits and the RV signal is a linear sum of Keplerians plus activity.
    Used throughout the fitting in Sec 5.

how reviews work

0 comments
Cite this review

Pith. "Pith review of High-Pass Filtering and Gaussian Process Regularization: Stellar Activity Characterization Techniques Applied to the 55 Cancri Planetary System." pith.science (2026). https://pith.science/paper/Z63HQ2M3

@misc{pith2026250908076,
  author       = {Pith},
  title        = {Pith review of: High-Pass Filtering and Gaussian Process Regularization: Stellar Activity Characterization Techniques Applied to the 55 Cancri Planetary System},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Z63HQ2M3}},
  note         = {Machine review of arXiv:2509.08076}
}
read the original abstract

Doppler planet searches are complicated by stellar activity, through which cyclical changes in the host star's photosphere and chromosphere can mask or mimic planetary signals. A popular technique for modeling stellar activity is to apply a quasiperiodic Gaussian process (GP) kernel, which provides a flexible model with rigorous error propagation. However, observers must guard against overfitting, as a GP may be flexible enough to subsume other signals besides the one it is intended to model. To counteract overfitting, we introduce a curvature-penalizing objective function for fitting GP models to long-term magnetic activity cycles. We also demonstrate that a Gaussian filter can be an effective method of detrending radial velocities (RVs) so that shorter-period signals can be extracted even in the absence of a mathematical model of the long-term trend. We apply our methods to the heavily studied 55 Cancri system, fitting Keplerian orbits plus the GP activity-cycle model. We show that a 4-Keplerian model that includes planets b, c, e, and f combined with a GP for the activity cycle performs at least as well as the widely agreed-upon 5-planet system with its own GP activity model. Our results suggest that the existence of planet d cannot be established from the RVs alone; additional data are required for confirmation.

Figures

Figures reproduced from arXiv: 2509.08076 by the authors.

Figure 1
Figure 1. Left: Long-term trend (dashed black line) computed using a Gaussian filter (GF) via Equation 2 superposed onto the 55 Cnc RV time series. The Gaussian filter used here has a width of σ = 500 days. The gray shaded region is comprised of 30 GF trends for a reasonable range of σ ∈ [400, 1000] days (sampled uniformly). We include a GF trend with a poorly chosen width of σ = 200 days (cyan dash-dotted line) for compariso… view at source ↗
Figure 2
Figure 2. Left: Periodograms of the undetrended (red) and detrended (blue) RVs, before removing planet b (left) and after removing planet b (right). The shaded gray rectangle has width 2R and shows the interval in which signals are not distinguishable from zero frequency. The black bracket has width 3R/4, which is the frequency separation of planet d’s orbit and the activity period given by Bourrier et al. (2018). vary in amp… view at source ↗
Figure 3
Figure 3. Periodograms for each residual RV during the Keplerian fitting process. At this stage, our fitting is done on the GF-detrended RVs. In order of removal, we fit Keplerians to planets b, c, f, and e. The horizontal dotted line represents the 1% false-alarm probability, as estimated by a 106 iteration bootstrap simulation of the periodogram. Here we describe how planet candidates are identified in Step 3 of the procedu… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Best-fit Keplerian orbits for planets b, c, f, and e. In the bottom right panel, we display the final RV residual after fitting these four planets in addition to the magnetic cycle. Note that the rightmost panels have time rather than orbital phase on the horizontal ax…
Figure 5
Figure 5. Figure 5: Features of the 5pGP model. Left: Phase-folded best-fit orbit of planet d. Center: Median GP prediction; note the dominance of signal drift over periodicity. Right: residual of the 5pGP model. Note the changing vertical axis scales. This plot uses the same color scheme…
Figure 6
Figure 6. Figure 6: Distribution of residuals normalized by the reported observation uncertainty. The best-fit Gaussian distributions are superposed. We interpret the distributions being broader than a standard normal distribution N (0, 1) as arising from jitter unaccounted for in the rep…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

106 extracted references · 19 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    Magnitude-squared Coherence: A Powerful Tool for Disentangling Doppler Planet Discoveries from Stellar Activity

    thebibliography [1] 20pt to REFERENCES 6pt =0pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command Each re...

  4. [4]

    2012, , 419, 3147, 10.1111/j.1365-2966.2011.19960.x

    Aigrain , S., Pont , F., & Zucker , S. 2012, , 419, 3147, 10.1111/j.1365-2966.2011.19960.x

  5. [5]

    J., Morales , J

    Alonso-Floriano , F. J., Morales , J. C., Caballero , J. A., et al. 2015, , 577, A128, 10.1051/0004-6361/201525803

  6. [6]

    W., & O'Neil , M

    Ambikasaran , S., Foreman-Mackey , D., Greengard , L., Hogg , D. W., & O'Neil , M. 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence, 38, 252, 10.1109/TPAMI.2015.2448083

  7. [7]

    W., & Darling, D

    Anderson, T. W., & Darling, D. A. 1952, The Annals of Mathematical Statistics, 23, 193. http://www.jstor.org/stable/2236446

  8. [8]

    1954, Journal of the American Statistical Association, 49, 765

    ---. 1954, Journal of the American Statistical Association, 49, 765. http://www.jstor.org/stable/2281537

Show all 106 references
  1. [9]

    2018, , 474, 2094, 10.1093/mnras/stx2109

    Angus , R., Morton , T., Aigrain , S., Foreman-Mackey , D., & Rajpaul , V. 2018, , 474, 2094, 10.1093/mnras/stx2109

  2. [10]

    P., Tollerud , E

    Astropy Collaboration , Robitaille , T. P., Tollerud , E. J., et al. 2013, , 558, A33, 10.1051/0004-6361/201322068

  3. [11]

    M., Sip o cz , B

    Astropy Collaboration , Price-Whelan , A. M., Sip o cz , B. M., et al. 2018, , 156, 123, 10.3847/1538-3881/aabc4f

  4. [12]

    M., Lim , P

    Astropy Collaboration , Price-Whelan , A. M., Lim , P. L., et al. 2022, , 935, 167, 10.3847/1538-4357/ac7c74

  5. [13]

    L., Donahue , R

    Baliunas , S. L., Donahue , R. A., Soon , W. H., et al. 1995, , 438, 269, 10.1086/175072

  6. [14]

    2024, , 167, 70, 10.3847/1538-3881/ad1330

    Beard , C., Robertson , P., Dai , F., et al. 2024, , 167, 70, 10.3847/1538-3881/ad1330

  7. [15]

    R., & Robinson , D

    Bevington , P. R., & Robinson , D. K. 2003, Data reduction and error analysis for the physical sciences

  8. [16]

    J., et al

    Blunt , S., Carvalho , A., David , T. J., et al. 2023, , 166, 62, 10.3847/1538-3881/acde78

  9. [17]

    2011, , 528, A4, 10.1051/0004-6361/201014354

    Boisse , I., Bouchy , F., H \'e brard , G., et al. 2011, , 528, A4, 10.1051/0004-6361/201014354

  10. [18]

    2021, , 161, 230, 10.3847/1538-3881/abec89

    Bortle , A., Fausey , H., Ji , J., et al. 2021, , 161, 230, 10.3847/1538-3881/abec89

  11. [19]

    2014, , 569, A65, 10.1051/0004-6361/201424266

    Bourrier , V., & H \'e brard , G. 2014, , 569, A65, 10.1051/0004-6361/201424266

  12. [20]

    2018, VizieR Online Data Catalog, J/A+A/615/A117

    Bourrier , V., Ehrenreich , D., Lecavelier Des Etangs , A., et al. 2018, VizieR Online Data Catalog, J/A+A/615/A117

  13. [21]

    G., Schmoll , J., Tyas , L

    Bramall , D. G., Schmoll , J., Tyas , L. M. G., et al. 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 8446, Ground-based and Airborne Instrumentation for Astronomy IV, ed. I. S. McLean , S. K. Ramsay , & H. Takami , 84460A, 10.1117/12.925935

  14. [22]

    2022, , 667, A68, 10.1051/0004-6361/202243167

    Bugnet , L. 2022, , 667, A68, 10.1051/0004-6361/202243167

  15. [23]

    P., Marcy , G

    Butler , R. P., Marcy , G. W., Williams , E., Hauser , H., & Shirts , P. 1997, , 474, L115, 10.1086/310444

  16. [24]

    P., Vogt , S

    Butler , R. P., Vogt , S. S., Laughlin , G., et al. 2017, , 153, 208, 10.3847/1538-3881/aa66ca

  17. [25]

    Cabot , S. H. C., Roettenbacher , R. M., Henry , G. W., et al. 2021, , 161, 26, 10.3847/1538-3881/abc41e

  18. [26]

    L., Handberg , R., Mathur , S., et al

    Campante , T. L., Handberg , R., Mathur , S., et al. 2011, , 534, A6, 10.1051/0004-6361/201116620

  19. [27]

    2014, , 567, A48, 10.1051/0004-6361/201323102

    Carolo , E., Desidera , S., Gratton , R., et al. 2014, , 567, A48, 10.1051/0004-6361/201323102

  20. [28]

    Cegla , H. M. 2019, Geosciences, 9, 114, 10.3390/geosciences9030114

  21. [29]

    N., & Rozelot , J

    Chowdhury , P., Kilcik , A., Yurchyshyn , V., Obridko , V. N., & Rozelot , J. P. 2019, , 294, 142, 10.1007/s11207-019-1530-7

  22. [30]

    2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Cosentino , R., Lovis , C., Pepe , F., et al. 2012, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 8446, Ground-based and Airborne Instrumentation for Astronomy IV, ed. I. S. McLean , S. K. Ramsay , & H. Takami , 84461V, 10.1117/12.925738

  23. [31]

    I., & Fabrycky , D

    Dawson , R. I., & Fabrycky , D. C. 2010, , 722, 937, 10.1088/0004-637X/722/1/937

  24. [32]

    B., S \'e gransan , D., Dumusque , X., et al

    Delisle , J. B., S \'e gransan , D., Dumusque , X., et al. 2018, , 614, A133, 10.1051/0004-6361/201732529

  25. [33]

    2011, in AAS/Division for Extreme Solar Systems Abstracts, Vol

    Demory , B.-O., Gillon , M., Deming , D., & Seager , S. 2011, in AAS/Division for Extreme Solar Systems Abstracts, Vol. 2, AAS/Division for Extreme Solar Systems Abstracts, 17.04

  26. [34]

    2016 a , , 455, 2018, 10.1093/mnras/stv2239

    Demory , B.-O., Gillon , M., Madhusudhan , N., & Queloz , D. 2016 a , , 455, 2018, 10.1093/mnras/stv2239

  27. [35]

    2016 b , , 532, 207, 10.1038/nature17169

    Demory , B.-O., Gillon , M., de Wit , J., et al. 2016 b , , 532, 207, 10.1038/nature17169

  28. [36]

    G., Endl , M., et al

    Desidera , S., Gratton , R. G., Endl , M., et al. 2004, , 420, L27, 10.1051/0004-6361:20040155

  29. [37]

    M., Galland , F., Udry , S., & Mayor , M

    Desort , M., Lagrange , A. M., Galland , F., Udry , S., & Mayor , M. 2007, , 473, 983, 10.1051/0004-6361:20078144

  30. [38]

    2024, , 167, 22, 10.3847/1538-3881/ad0c58

    Dodson-Robinson , S., & Haley , C. 2024, , 167, 22, 10.3847/1538-3881/ad0c58

  31. [39]

    E., Delgado , V

    Dodson-Robinson , S. E., Delgado , V. R., Harrell , J., & Haley , C. L. 2022, , 163, 169, 10.3847/1538-3881/ac52ed

  32. [40]

    2012, PhD thesis, University of Geneva, Astronomical Observatory

    Dumusque , X. 2012, PhD thesis, University of Geneva, Astronomical Observatory

  33. [41]

    C., & Monteiro , M

    Dumusque , X., Udry , S., Lovis , C., Santos , N. C., & Monteiro , M. J. P. F. G. 2011, , 525, A140, 10.1051/0004-6361/201014097

  34. [42]

    2017, , 598, A133, 10.1051/0004-6361/201628671

    Dumusque , X., Borsa , F., Damasso , M., et al. 2017, , 598, A133, 10.1051/0004-6361/201628671

  35. [43]

    2012, , 547, A18, 10.1051/0004-6361/201219981

    Ehrenreich , D., Bourrier , V., Bonfils , X., et al. 2012, , 547, A18, 10.1051/0004-6361/201219981

  36. [44]

    A., El-Taher , A

    El-Borie , M. A., El-Taher , A. M., Thabet , A. A., et al. 2020, , 898, 73, 10.3847/1538-4357/ab9d21

  37. [45]

    D., et al

    Endl , M., Robertson , P., Cochran , W. D., et al. 2012, , 759, 19, 10.1088/0004-637X/759/1/19

  38. [46]

    J., de Mooij , E

    Esteves , L. J., de Mooij , E. J. W., Jayawardhana , R., Watson , C., & de Kok , R. 2017, , 153, 268, 10.3847/1538-3881/aa7133

  39. [47]

    M., Pope , B

    Farr , W. M., Pope , B. J. S., Davies , G. R., et al. 2018, , 865, L20, 10.3847/2041-8213/aadfde

  40. [48]

    A., Marcy , G

    Fischer , D. A., Marcy , G. W., Butler , R. P., et al. 2008, , 675, 790, 10.1086/525512

  41. [49]

    A., Anglada-Escude , G., Arriagada , P., et al

    Fischer , D. A., Anglada-Escude , G., Arriagada , P., et al. 2016, , 128, 066001, 10.1088/1538-3873/128/964/066001

  42. [50]

    2017, The Astronomical Journal, 154, 220, 10.3847/1538-3881/aa9332

    Foreman-Mackey, D., Agol, E., Ambikasaran, S., & Angus, R. 2017, The Astronomical Journal, 154, 220, 10.3847/1538-3881/aa9332

  43. [52]

    2023 b , , 672, A149, 10.1051/0004-6361/202245201

    ---. 2023 b , , 672, A149, 10.1051/0004-6361/202245201

  44. [53]

    2012, Computational Optimization and Applications, 51, 259 , 10.1007/s10589-010-9329-3

    Gao, F., & Han, L. 2012, Computational Optimization and Applications, 51, 259 , 10.1007/s10589-010-9329-3

  45. [54]

    1972, The analysis of tides

    Godin , G. 1972, The analysis of tides. (Toronto: Univ. of Toronto Press)

  46. [55]

    C., Bonfils , X., et al

    Gomes da Silva , J., Santos , N. C., Bonfils , X., et al. 2012, , 541, A9, 10.1051/0004-6361/201118598

  47. [56]

    Gray, D. F. 2009, The Astrophysical Journal, 697, 1032, 10.1088/0004-637X/697/2/1032

  48. [57]

    1994, Nonparametric regression and generalized linear models: a roughness penalty approach (United Kingdom: Chapman and Hall)

    Green, P., & Silverman, B. 1994, Nonparametric regression and generalized linear models: a roughness penalty approach (United Kingdom: Chapman and Hall)

  49. [58]

    2024, 55Cnc\_exoplanets, v1.0.0, Zenodo, 10.5281/zenodo.14571275

    Harrell, J. 2024, 55Cnc\_exoplanets, v1.0.0, Zenodo, 10.5281/zenodo.14571275

  50. [59]

    Hathaway, D. H. 2015, Living Reviews in Solar Physics, 12, 4, 10.1007/lrsp-2015-4

  51. [60]

    Hatzes , A. P. 2002, Astronomische Nachrichten, 323, 392, 10.1002/1521-3994(200208)323:3/4<392::AID-ASNA392>3.0.CO;2-M

  52. [61]

    Hatzes, A. P. 2019, The Doppler Method for the Detection of Exoplanets, 2514-3433 (IOP Publishing), 10.1088/2514-3433/ab46a3

  53. [62]

    D., Collier Cameron , A., Queloz , D., et al

    Haywood , R. D., Collier Cameron , A., Queloz , D., et al. 2014, , 443, 2517, 10.1093/mnras/stu1320

  54. [63]

    L., et al

    Hinshaw , G., Barnes , C., Bennett , C. L., et al. 2003, , 148, 63, 10.1086/377222

  55. [64]

    J., & Prescott, E

    Hodrick, R. J., & Prescott, E. C. 1997, Journal of Money, Credit and Banking, 29, 1. http://www.jstor.org/stable/2953682

  56. [65]

    E., & Raftery, A

    Kass, R. E., & Raftery, A. E. 1995, Journal of the American Statistical Association, 90, 773, 10.1080/01621459.1995.10476572

  57. [66]

    2019, , 623, A72, 10.1051/0004-6361/201834371

    Kervella , P., Arenou , F., Mignard , F., & Th \'e venin , F. 2019, , 623, A72, 10.1051/0004-6361/201834371

  58. [67]

    L \'o pez-Morales , M., Triaud , A. H. M. J., Rodler , F., et al. 2014, , 792, L31, 10.1088/2041-8205/792/2/L31

  59. [68]

    2010, in Exoplanets, ed

    Lovis , C., & Fischer , D. 2010, in Exoplanets, ed. S. Seager (University of Arizona Press), 27--53

  60. [69]

    2021, , 162, 61, 10.3847/1538-3881/ac0057

    Lubin , J., Robertson , P., Stefansson , G., et al. 2021, , 162, 61, 10.3847/1538-3881/ac0057

  61. [70]

    W., Butler , R

    Marcy , G. W., Butler , R. P., Fischer , D. A., et al. 2002, , 581, 1375, 10.1086/344298

  62. [71]

    E., Endl , M., Cochran , W

    McArthur , B. E., Endl , M., Cochran , W. D., et al. 2004, , 614, L81, 10.1086/425561

  63. [72]

    1999, IEEE Signal Processing Magazine, 16, 29, 10.1109/MSP.1999.790977

    McClellan, J., & Schafer, R. 1999, IEEE Signal Processing Magazine, 16, 29, 10.1109/MSP.1999.790977

  64. [73]

    Meunier , N., Desort , M., & Lagrange , A. M. 2010, , 512, A39, 10.1051/0004-6361/200913551

  65. [74]

    2013, Astronomische Nachrichten, 334, 141, https://doi.org/10.1002/asna.201211742

    Meunier, N., & Lagrange, A. 2013, Astronomische Nachrichten, 334, 141, https://doi.org/10.1002/asna.201211742

  66. [75]

    A., & Mead, R

    Nelder, J. A., & Mead, R. 1965, The Computer Journal, 7, 308, 10.1093/comjnl/7.4.308

  67. [76]

    R., Irwin, J., Charbonneau, D., Berta-Thompson, Z

    Newton, E. R., Irwin, J., Charbonneau, D., Berta-Thompson, Z. K., & Dittmann, J. A. 2016, The Astrophysical Journal Letters, 821, L19, 10.3847/2041-8205/821/1/L19

  68. [77]

    1996, The VizieR database of astronomical catalogues, CDS, Centre de Données astronomiques de Strasbourg, 10.26093/CDS/VIZIER

    Ochsenbein, F. 1996, The VizieR database of astronomical catalogues, CDS, Centre de Données astronomiques de Strasbourg, 10.26093/CDS/VIZIER

  69. [78]

    2000, , 143, 23, 10.1051/aas:2000169

    Ochsenbein , F., Bauer , P., & Marcout , J. 2000, , 143, 23, 10.1051/aas:2000169

  70. [79]

    2003, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Pepe , F., Rupprecht , G., Avila , G., et al. 2003, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4841, Instrument Design and Performance for Optical/Infrared Ground-based Telescopes, ed. M. Iye & A. F. M. Moorwood , 1045--1056, 10.1117/12.460777

  71. [80]

    2008, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Perruchot , S., Kohler , D., Bouchy , F., et al. 2008, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 7014, Ground-based and Airborne Instrumentation for Astronomy II, ed. I. S. McLean & M. M. Casali , 70140J, 10.1117/12.787379

  72. [81]

    \'A ., & Lindegren , L

    Perryman , M., Hartman , J., Bakos , G. \'A ., & Lindegren , L. 2014, , 797, 14, 10.1088/0004-637X/797/1/14

  73. [82]

    A., Reece , S., & Roberts , S

    Rajpaul , V., Aigrain , S., Osborne , M. A., Reece , S., & Roberts , S. 2015, , 452, 2269, 10.1093/mnras/stv1428

  74. [83]

    M., Buchhave , L

    Rajpaul , V. M., Buchhave , L. A., Lacedelli , G., et al. 2021, , 507, 1847, 10.1093/mnras/stab2192

  75. [84]

    S., Dodson-Robinson , S., & Haley , C

    Ramirez Delgado , V., Caicedo Vivas , J. S., Dodson-Robinson , S., & Haley , C. 2025, arXiv e-prints, arXiv:2506.20864, 10.48550/arXiv.2506.20864

  76. [85]

    E., & Williams, C

    Rasmussen, C. E., & Williams, C. K. I. 2005, Gaussian Processes for Machine Learning (The MIT Press), 10.7551/mitpress/3206.001.0001

  77. [86]

    O., & Uhlig, H

    Ravn, M. O., & Uhlig, H. 2002, The Review of Economics and Statistics, 84, 371, 10.1162/003465302317411604

  78. [87]

    R., Snellen , I

    Ridden-Harper , A. R., Snellen , I. A. G., Keller , C. U., et al. 2016, , 593, A129, 10.1051/0004-6361/201628448

  79. [89]

    2013 b , , 764, 3, 10.1088/0004-637X/764/1/3

    ---. 2013 b , , 764, 3, 10.1088/0004-637X/764/1/3

  80. [90]

    2014, The Astrophysical Journal Letters, 793, L24, 10.1088/2041-8205/793/2/L24

    Robertson, P., & Mahadevan, S. 2014, The Astrophysical Journal Letters, 793, L24, 10.1088/2041-8205/793/2/L24

  81. [91]

    2014, Science, 345, 440, 10.1126/science.1253253

    Robertson , P., Mahadevan , S., Endl , M., & Roy , A. 2014, Science, 345, 440, 10.1126/science.1253253

  82. [92]

    H., & Donahue , R

    Saar , S. H., & Donahue , R. A. 1997, , 485, 319, 10.1086/304392

  83. [93]

    W., & Stephens, M

    Scholz, F. W., & Stephens, M. A. 1987, Journal of the American Statistical Association, 82, 918. http://www.jstor.org/stable/2288805

  84. [94]

    1978, Annals of Statistics, 6, 461

    Schwarz , G. 1978, Annals of Statistics, 6, 461

  85. [95]

    H., & Stoffer, D

    Shumway, R. H., & Stoffer, D. S. 2001, Time Series Analysis and Its Applications, 4th edn. (New York: Springer)

  86. [96]

    Su \'a rez Mascare \ n o , A., Rebolo , R., & Gonz \'a lez Hern \'a ndez , J. I. 2016, , 595, A12, 10.1051/0004-6361/201628586

  87. [97]

    I., & Esposito , M

    Su \'a rez Mascare \ n o , A., Rebolo , R., Gonz \'a lez Hern \'a ndez , J. I., & Esposito , M. 2017, , 468, 4772, 10.1093/mnras/stx771

  88. [98]

    R., et al

    Su \'a rez Mascare \ n o , A., Gonz \'a lez- \'A lvarez , E., Zapatero Osorio , M. R., et al. 2023, , 670, A5, 10.1051/0004-6361/202244991

  89. [99]

    J., & Haley , C

    Thomson , D. J., & Haley , C. L. 2014, Proceedings of the Royal Society of London Series A, 470, 20140101, 10.1098/rspa.2014.0101

  90. [100]

    G., MacQueen , P

    Tull , R. G., MacQueen , P. J., Sneden , C., & Lambert , D. L. 1995, , 107, 251, 10.1086/133548

  91. [101]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261, 10.1038/s41592-019-0686-2

  92. [102]

    Vogt , S. S. 1987, , 99, 1214, 10.1086/132107

  93. [103]

    S., Allen , S

    Vogt , S. S., Allen , S. L., Bigelow , B. C., et al. 1994, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 2198, Instrumentation in Astronomy VIII, ed. D. L. Crawford & E. R. Craine , 362, 10.1117/12.176725

  94. [104]

    S., ten Brummelaar , T

    von Braun , K., Boyajian , T. S., ten Brummelaar , T. A., et al. 2011, , 740, 49, 10.1088/0004-637X/740/1/49

  95. [105]

    N., Matthews , J

    Winn , J. N., Matthews , J. M., Dawson , R. I., et al. 2011, , 737, L18, 10.1088/2041-8205/737/1/L18

  96. [106]

    2005, in AAS/Division of Dynamical Astronomy Meeting, Vol

    Wisdom , J. 2005, in AAS/Division of Dynamical Astronomy Meeting, Vol. 36, AAS/Division of Dynamical Astronomy Meeting \#36, 05.08

  97. [107]

    Wright , J. T. 2004, , 128, 1273, 10.1086/423221

  98. [109]

    2023, Nature Astronomy, 7, 198, 10.1038/s41550-022-01837-2

    ---. 2023, Nature Astronomy, 7, 198, 10.1038/s41550-022-01837-2

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.