One-Way Analysis of Variance with Time-Series Data

Abstract

SUMMARY Successive observations over time are made on individual subjects classified into different groups. It is assumed that the mean response may vary between groups, that there is a random effect for each individual, and that successive observations on each individual follow a simple ARMA model. The problem of testing the null hypothesis of equality of the group means is considered. In many practical situations the usual analysis-of-variance F test, performed on averages of the observations over time, provides an efficient test.