χ2 Discriminators for Transiting Planet Detection in Kepler Data

The Kepler spacecraft observes a host of target stars to detect transiting planets. Requiring a 7.1σ detection in three years of data yields over 100,000 detections, many of which are false alarms. After a second cut is made on a robust detection statistic, some 50,000 or more targets still remain. These false alarms waste resources as they propagate through the remainder of the software pipeline and so a method to discriminate against them is crucial in maintaining the desired sensitivity to true events. This paper describes a χ² test which represents a novel application of an existing formalism developed for false alarm mitigation in searches for gravitational waves. Using this technique, the false alarm rate can be lowered to ~5%.