2016MNRAS.461..627S

An automated tool to derive the off-pulse interval of a light curve originating from a pulsar is needed. First, we derive a powerful and accurate non-parametric sequential estimation technique to estimate the off-pulse interval of a pulsar light curve in an objective manner. This is in contrast to the subjective 'eye-ball' (visual) technique, and complementary to the Bayesian Block method which is currently used in the literature. The second aim involves the development of a statistical package, necessary for the implementation of our new estimation technique. We develop a statistical procedure to estimate the off-pulse interval in the presence of noise. It is based on a sequential application of p-values obtained from goodness-of-fit tests for uniformity. The Kolmogorov-Smirnov, Cramer-von Mises, Anderson-Darling and Rayleigh test statistics are applied. The details of the newly developed statistical package sopie (Sequential Off-Pulse Interval Estimation) are discussed. The developed estimation procedure is applied to simulated and real pulsar data. Finally, the sopie estimated off-pulse intervals of two pulsars are compared to the estimates obtained with the Bayesian Block method and yield very satisfactory results. We provide the code to implement the sopie package, which is publicly available at http://CRAN.R-project.org/package=SOPIE (Schutte).