MATHEMATICAL TREATMENT OF RANDOM PROCESSES WITH DEAD TIME.

Random events measured with a constant dead time give a counting rate smaller than the event rate. The latter has been obtained after dead time correction. A current method of dead time correction is proved to be justified only if the event rate is constant in time. In general the average event rate is obtained by dividing the average number of counts by the average length of live time, the measuring time minus dead time. The probability distribution and variance are explicitly given. The present work has been motivated by the data analysis of a rocket experiment of fast electrons quasi-trapped in the geomagnetic field and measured with proportional counters.