Statistical time analysis for regular events with high count rate

In physics, it is frequently needed to precisely measure the count rate of some process. Quite often one needs to account for electronics dead time, pile-up and other features of data acquisition system to avoid systematic shifts of the count rate. In this article, we present a statistical mechanism to diminish or completely eliminate systematic errors arising from the correlation between the events. Also, we present examples of application of this method to the analysis of "Troitsk nu-mass" and "Tristan in Troitsk" experiments.