Using mean reversion as a measure of persistence

This paper elaborates on the alternative measure of persistence recently suggested in Marques (2004), which is based on the idea of mean reversion. A formal distinction between the “unconditional probability of a given process not crossing its mean in period t” and its estimator, is made clear and the relationship between this new measure and the widely used “sum of the autoregressive coefficients”, as alternative measures of persistence, is investigated. Using the law of large numbers and the central limit theorem, properties for the estimator of the new measure of persistence are established, which allow tests of hypotheses to be performed, under very general conditions. Finally, some Monte Carlo experiments are conducted in order to compare the finite sample properties of the estimator for the “unconditional probability of a given process not crossing its mean in period t” and the OLS estimator for the “sum of the autoregressive coefficients”.