A two-mode prediction test has low sensitivity to an imposed 30 ppm signal in KIC 10001893 residuals

We test whether two known stellar pulsations can predict the known 14.261 microHz signal in KIC 10001893. A fixed complex response uses the product of parent coefficients at 210.6816 and 196.4174 microHz. We fit the response in Kepler release 25 quarters 3 to 10 and test quarters 11 to 17. For 60-day windows, the response increases squared prediction error by 23 percent in aperture flux and 9 percent in corrected flux, relative to a steady signal. Quarter resampling gives wide intervals that include a gain. The weak parent has local coefficient errors near 45 to 50 ppm. In 500 residual tests with an imposed quadratic signal at a 30 ppm reference amplitude, measured parents give a positive gain in only 37 and 27 trials. True parent coefficients give a gain in 412 and 406 trials. Thus this test has low sensitivity under these noise and signal assumptions. The observed null result does not reject nonlinear pulsation or confirm a planet. The two flux products share exposures. Code, input hashes, local fits, aggregate results and external retrieval steps are included.

astronomy · Kepler · KIC 10001893 · difference frequencies · nonlinear pulsation · prediction · sensitivity