Practical properties of the CUSUM process
We establish new properties of the Lindley process, also known as the waiting time process or the cumulative-sum (CUSUM) process, as well as its running maximum. Unlike many other authors, the study focuses on exact and asymptotic expressions for the moment generating function (MGF) of CUSUM. It includes precise finite-sample expressions for the MGF and moments of the process, along with fast recursive computing algorithms, lower and upper bounds, and the classification of large-sample asymptotics of the CUSUM MGF. Results are applied to single, multiple, and transient change-point problems, for the calculation of thresholds that provide a desired control of familywise false alarm rates, as well as the quantiles of queuing processes and probabilities of their large deviation at least once over a given finite time interval.