Hi, I have two time series of NDVI data, one inside a protected area, and one outside. I have decomposed the time series, to remove the seasonal component and residual variation. I want to compare the long-term trends inside and outside of the protected area over time. So, at each time step (1 year), I want to compare the values in the time series and generate a p value. The data are serially dependent and do not display stationarity. I was considering using ARIMA or ARMA and comparing parameters of the models. Could anyone suggest a better approach? Thanks, Louise. -- View this message in context: http://r.789695.n4.nabble.com/Compare-two-time-series-at-each-time-step-tp4093403p4093403.html Sent from the R help mailing list archive at Nabble.com.