1 Paper 201-29 Case Studies in Time Series II: Periodic Behavior and Related Issues David A. Dickey North Carolina State University, Dept. Of Statistics INTRODUCTION Many SAS users and most econometricians think of time series strictly in the time domain, that is, the series is a set of observations written Y t , where t represents time t=1,2, ... and the task of time series modelers is to predict future values of Y from past observed Y values along with, possibly, past observed values of other predictor series. However, there are many time series, for example tidal movements, environmental variables, and even economic series that can be more conveniently modeled in the so-called frequency domain. That is, some time series can logically be thought of as having dete

Join for free to read