Estimated error scales excluded 25 of 45 Kepler sine selections when outliers were kept; 16 had larger later error
We tested two early selection rules for later Kepler transit times. One rule uses the stated timing errors as fixed scales. The other allows an unknown common error scale for each planet. A sine cycle was set from early orbital periods of timed neighbours near a first-order period ratio. Models used one shared cutoff per star near 70% of the recorded span. With source outlier rows kept and other quality rules fixed, 92 planets were eligible. The fixed scale rule selected 45 planets. The estimated scale rule selected 20 of them. Of the 25 excluded selections, 16 had larger later weighted root mean square error than a line, and 9 had lower error. The estimated scale selections improved 16 of 20 forecasts. Under strict source quality filters, that rule selected 23 of 87 planets; 21 had lower later error. Their median sine-to-line error ratio was 0.806, with a 95% star bootstrap interval of 0.710 to 0.895. Paired tests of the complete selection policies did not establish lower overall error for either rule. The matched rule comparison was added after earlier results were seen. It is exploratory. Source templates, flags, and status use later catalogue information. The test is retrospective and concerns known planets. The package preserves both numerical inputs, provenance, original code, all eligible scores, predictions, figures, and offline reproduction instructions.
Kepler · transit timing · orbital periods · forecast validation · measurement errors · model selection · public astronomical data
References
8- usesLater quadratic timing errors were larger than linear errors for 16 of 28 selected Kepler objects with a 70/30 split sci:sha256:da64194e286627a311d45cba907c643cfba74dd47966c96b347c71fb747f32c5
We reuse its frozen NASA status table and source log. We wrote new analysis code. Its quadratic forecast test helped set the initial question.
- supportsAn orbital beat model reduced later timing error for 25 of 26 Kepler planets selected from filtered early data sci:sha256:31a304247cf62293691375e5a9ec45f72a4c37a2c6d2ae0bec5d3aa4aba351a7
Our different shared cutoff test also finds lower later error for most early-selected orbital sine models. After reading this work, we added a matched comparison of error scale rules. We do not reuse its fits or code.
- usesdoi:10.3847/0067-0049/225/1/9
Table 3 supplies the measured transit offsets, errors, expected times, and quality flags.
- usesdoi:10.26093/cds/vizier.22250009
The bundled timing input is the unmodified numeric CDS table3.dat.gz file.
- usesdoi:10.26133/NEA4
The frozen NASA KOI table supplies star numbers, planet names, and combined status. NASA periods do not enter the calculation.
- usesdoi:10.1088/0004-637X/761/2/122
The near-resonance timing cycle sets the fixed sine cycle from early periods.
- doi:10.3847/1538-4365/aa9f2b
Prior work used sine timing models in Kepler photometry. The model family is established.
- doi:10.1093/mnras/staf334
Prior work used later TESS times to assess timing predictions for Kepler and K2 planets.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- Sequential drift fits gave less timing error than rolling lines for 38 of 49 selected Kepler objects
Prior Kepler prediction work tests estimated error scales and quality-dependent selection for fixed orbital-beat sine forecasts.
- usesOrbital beat periods gave lower mean later error than fitted periods in 146 Kepler records without model selection
Its test of early selection and error-scale assumptions informed our conditional score and common-cutoff check. Our analysis code is independent.
- No added circular chopping term passed early selection in 146 Kepler timing records
Related SCI work compares timing selection rules with estimated and fixed error scales.