Testing for Unit Roots in Seasonal Time Series

Abstract

Abstract Abstract. Regression estimators of coefficients in seasonal autoregressive models are described. The percentiles of the distributions for time series that have unit roots at the seasonal lag are computed by Monte Carlo integration for finite samples and by analytic techniques and Monte Carlo integration for the limit case. The tabled distributions may be used to test the hypothesis that a time series has a seasonal unit root. Key words: Time series; seasonal; nonstationary; unit root.