On Tue, 28 Jan 2003 23:42:49 GMT, "Mountain Bikn' Guy" <[EMAIL PROTECTED]> wrote: < responding to me, where I asked ,> > > X is autocorrelated, and Y is autocorrelated?
> Y is autocorrelated. The X variables are of many different data types, > including nominal, ordinal, interval and continuous. Some of the continuous > X's will show autocorrelation. Actually, our data set has thousands of > independent variables, and I have not actually tested every one for > autocorrelation. You might get some advice that is usefully tuned to your concerns, if you reveal more about the data. How might those nominal variables relate to the time series? I think of my nominal variables as each keeping one value for a long period of time. That would always be the case for the 'time series' that can be subdivided into smaller intervals, like those series on temperatures, etc. If the nominal has each value for a while, then the Time series can be (and ought to be) aggregated into equivalent units of time -- if there is a *big*, overall main effect. On the other hand, if the nominal variable has an influence that is 'temporary' as the times are scaled, then the proper analysis only uses data from the time when the nominal *changes*. There are several things that can be called 'event analyses'. It might be that for any given change from B to C, there could be a systematic change in the time series, over some fairly short sequence. But it would be stupid to test all those possibilities, with thousands of variables and millions of time-frames, if there wasn't a logical basis for some question. -- Rich Ulrich, [EMAIL PROTECTED] http://www.pitt.edu/~wpilib/index.html . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
