sciscape Publications
← All publications
astro-ph.EP · Earth and Planetary Astrophysicsastro-ph.IM · Instrumentation and Methods for Astrophysics

Corrected Kepler DR25 point fits give AUC 0.517 for true b ≥ 0.8 in selected recoveries at expected MES 6–20

We test how well supplied fitted impact parameters identify high-impact transits in public Kepler DR25 pixel injections. The sample has 25,546 recovered circular injections with corrected supplemental fits. Stars have temperature 4000–7000 K and log surface gravity above 4. Expected multiple event statistic (MES) is 6–20. Period is below 350 days. Duration is 1.5 to below 15 hours. The source transit count is at least four. Injected planet-to-star radius ratio is at most 0.1. High impact means true b ≥ 0.8. The area under the receiver operating characteristic curve (AUC) for fitted b is 0.5172. Its 95% interval is 0.5047–0.5291 from 2000 draws of whole source sky groups. The interval is below the specified practical bound of 0.60. Thresholds tested on five held-out sky-group folds remove 7.47% of high-impact cases while removing 4.96% of low-impact cases. Excess removal is 2.50 percentage points, with interval 1.34–3.66. That interval holds trained thresholds fixed. A separate fitted b < 0.9 cut rejects 1424 of 24598 true passes and retains 846 of 948 true failures. We extend prior fixed-cut tests with full threshold curves, AUC values by expected MES, and held-out decisions. The result describes the selected synthetic recoveries and supplied point fits. It gives no real-planet error probability and does not test catalog posterior fits. The package preserves both source responses, raw hashes, code, row decisions, figures, independent audits, and offline repeat instructions.

Kepler DR25 · transit injection · impact parameter · geometry selection · receiver operating characteristic · AUC · held-out validation · reproducible astronomy

Admitted 2 citing publications

References

11
  1. usesAdding the long CDPP slope reduces Brier score in a fixed Kepler DR25 recovery model for FGK dwarfs at expected MES 6–20 sci:sha256:4ada080bad333fe183c9c73d094f32590390941a9024b289f572b15ac74b4899

    We use its frozen NASA source responses and confirm their hashes with fresh direct downloads.

  2. usesSelected supplemental Kepler DR25 fits have a median radius ratio error of −11.1% at injected b ≥ 0.8 and expected MES 8–20 sci:sha256:0182a689c1dae0e0891318239c1fffdb9d8c2c0abc6eb2db1d23915b77ecf67e

    Its fitted b < 0.8 selection table motivates this extension across all thresholds and to held-out sky groups.

  3. supportsKepler DR25 supplemental fits give median radius-ratio errors of +4.1% at input b<0.5 and −10.2% at b≥0.8 in selected recoveries with measured MES 7.1–10 sci:sha256:0e26edce17b1d791e769d2bf5ace374de32ebdcd01d122db5798879b5a64f4ea

    The weak geometry separation agrees with its related findings for corrected fit errors and mean fitted impact parameter.

  4. Circular density errors exceed a factor of two in 39% of selected weak Kepler DR25 fits with tabulated duration errors at most 10% sci:sha256:f9b8ea07cfb68377e597ac1c57b80bbdc1ac66589c66334508168c30e1fc3331

    This related study tests geometry and circular-density errors when duration agrees.

  5. usesdoi:10.26133/NEA14

    This is the source Kepler completeness and reliability data product.

  6. usesdoi:10.26133/NEA6

    We use the DR25 stellar table to select the target stars.

  7. usesdoi:10.26133/NEA34

    The guide defines the injection table and the correction that selects supplemental fits.

  8. usesdoi:10.3847/1538-4365/229/2/30

    This paper describes the DR25 stellar properties used for selection.

  9. usesdoi:10.3847/1538-3881/abab0b

    This paper describes the pixel injection experiment and pipeline recovery.

  10. doi:10.3847/1538-3881/ab9fff

    This prior work explains weak geometry constraints in long-cadence data and a duration-based selection method.

  11. doi:10.3847/1538-3881/ac45f4

    This prior work studies transit geometry degeneracy and improved posterior sampling.

Cited by

2 publications

Public catalog entries on Sciscape.