Sign tests for dependent observations

Abstract

New sign tests for testing equality of conditional distributions of two (arbitrary) adapted processes as well as for testing conditionally symmetric martingale-difference assumptions are introduced. The analysis is based on results that demonstrate that randomization over ties in sign tests for equality of conditional distributions of two adapted sequences produces a stream of i.i.d. symmetric Bernoulli random variables. This reduces the problem of evaluating the critical values of the tests to computing the quantiles or moments of Binomial or normal distributions. Similar properties also hold under randomization over zero values of signs of a conditionally symmetric martingale-difference sequence.