The previous transit reduced squared prediction error by 4.5% for 769 later Kepler-1520 b transits with fixed rotation terms
We tested whether the measured light loss in the previous transit improves prediction of the next transit of Kepler-1520 b. We measured mean net light loss for 1,814 transits from 17 public Kepler DR25 light curves. We used a fixed historical DR24 ephemeris. Two linear rules were fitted on 684 eligible transits in quarters 1 through 8. Their coefficients stayed fixed for 769 test transits in quarters 9 through 17. Both rules used a recent mean and fixed harmonics at a 22.9 day rotation period. Adding the previous loss minus the recent mean reduced mean squared error by 4.51%. Root mean squared error fell from 625.58 to 611.32 parts per million. A conditional 95% interval from blocks of 35 orbits was 0.93% to 8.08%. The gain was positive in seven of nine test quarters. It was similar with corrected flux, a quadratic local baseline, and almost complete measurement windows. The model without rotation terms had lower absolute error, but the interval for its gain included zero. The interval also included zero after quiet times were excluded. These results apply to the stated rules and data split. They do not establish a dust grain lifetime or a physical cause. The package preserves all source light curves, source hashes, measurements, individual predictions, code, and figures. A fresh run reproduced all generated result and figure files exactly in the recorded environment.
Kepler-1520 b · KIC 12557548 · disintegrating planets · transit variability · time series prediction · public astronomical data · reproducible analysis
References
5- doi:10.1088/0004-637X/752/1/1
This paper reports the variable transits and the proposed dusty outflow from the planet.
- doi:10.1051/0004-6361/201322398
This paper studies transit depth statistics in 15 Kepler quarters. Its measurement and correlation test differ from our prediction test.
- usesdoi:10.1093/mnras/stv297
We use the reported 22.9 day rotation period to set the fixed harmonic inputs. We reuse no measured depths or code.
- doi:10.1051/0004-6361/201629250
This paper gives physical dust models and context for the limits of a dust lifetime interpretation.
- A free tail amplitude gives no stable gain in gap tests on 22 BD+05 4868 Ab transits from TESS sectors 55 and 82 sci:sha256:9ebf45a2e22ca48f2f397ac100cf834f0117883331528b91bcbf06f063153ea6
This related benchmark predicts missing parts within observed transits of another disintegrating planet. Our test predicts the next transit from earlier measurements. We reuse no code or data.
Cited by
9 publicationsPublic catalog entries on Sciscape.
- usesA fixed profile test does not detect a tail response after one orbit in Kepler-1520 b quarters 9–17
We reuse its preserved Kepler FITS files, source hashes, archive URLs, and fixed DR24 ephemeris. Its broad transit prediction result motivates this separate core and tail response test. We do not reuse its analysis code.
- A reference shift gave a lower 100-day geometric mean timing error than a full refit after three late times in 28 Kepler records with early correlation above 0.5
This related sequential prediction study used the previous Kepler-1520 b transit to predict light loss. Its target differs from timing.
- usesThree past flux inputs increased squared error by 0.83% in a fixed linear test of 522 Kepler-1520 b transits
We reuse its verified raw Kepler files, source hashes, fixed archive ephemeris, unchanged measurement code, history inputs, and training-only least-squares routine. We add past-only local harmonic flux inputs and a separate paired prediction test.
- usesSix 2024 TESS sectors show mean dimming at the fixed Kepler-1520 b transit phase
We reuse its public Kepler source files and hashes, fixed historical ephemeris choice, and broad measurement window. We make new measurements and test six TESS sectors from 2024.
- The primary Kepler Q1 to Q17 test leaves adjacent-orbit memory in KOI-2700 b uncertain under matched noise
Its catalog abstract tests transit prediction for Kepler-1520 b, a different object. It gives related context. We do not reuse its code or results.
- Fitted KOI-2700b transit changes exceed control changes in Kepler Q1 to Q17 after windows with attitude tweaks are removed
Related transit-change research on Kepler-1520b. The target and statistic differ. No code or results are reused.
- usesA fit with a fixed template shows mean aperture dimming in four combined 2022 TESS sectors of Kepler-1520
We reuse its NASA Exoplanet Archive DR24 ephemeris record and provenance.
- usesLocal stellar rotation terms did not reduce prediction error for 593 later Kepler-1520 b transits in a fixed linear test
We reuse its source data, fixed ephemeris and unchanged MIT code for transit measurement, history inputs, OLS and conditional block ranges.
- usesLocal flux inputs reduced squared test error by 0.90% for 769 Kepler-1520 b transits with shared linear baselines
We reuse its verified native Kepler data, ephemeris, measurement code, history inputs, least-squares routine and block-resampling routine. We reproduce its three measurement series exactly and test added stellar flux inputs.