TTVs are derived from Q0-Q17 Kepler data. x-axes: “Observed Tc” (Mid-Transit Time): EXOFAST’s best-fits from Normalized PDCSAP_FLUX Kepler light flux vs. time (BJD_tdb - 2454900) data. y-axes: “(O-C)”: difference between Observed Tc and the Calculated Tc from the graphically obtained linear ephemeris.
Figure: KOI-198.01, P = 87.24 days [Plot avg. error bars = ± 0.53 min. (smaller than symbols)]
Linear ephemeris (this work): Tc = 87.24223896(Tc#) + 86.35033475
* https://cfop.ipac.caltech.eduLiterature Tc#, Tc, and TTV Major Tabulations and Other References: see my "Summary" webpage.
16 Sep 2014
Kepler KOI-198 (KIC-10666242) 2-(or more?)-Planet System
The presence of both a credible (i.e.: False Alarm Probability (FAP) < 0.1) periodicity in the Lomb-Scargle Periodogram (LSP) of [(O-C) vs. Time] ("TTVxy") data, along with a sinusoidal distribution of the data displaying a corresponding periodicity ("P_ttv"), is strongly suggestive of the presence of an unseen (non-transiting), gravitationally perturbing planetary object(s) in the system. However, when periodicity values from both methods are large and distinctly visually obvious, they may not correspond to each other very closely if considerably less than one cycle of periodicity has been observed. This condition would significantly affect both the choice of "best-fit" sinusoidal curve and the precision of the LSP periodicity…and is apparently the case for KOI-198.01 (figure and data follow).
Note also that the best sinusoidal curve-fit is not a perfect match for the data in the following figure suggesting that this system is not made up of strictly circular orbits. Furthermore, while 198.01 is currently listed on the Kepler Community Follow-up Observing Program (CFOP) website* as a planetary candidate, a note is included there that it may indeed be part of an eclipsing binary. Unless the K2 project can shed more light on this uncertainty, it may remain such for some time to come.