[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: RE: Residuals after panel estimation
In addition, see the -modeldiag- package, which may help.
-search- in an up-to-date Stata will point to files.
> But xtline is a command in Stata 8 to do line plots with panel data !
> Would that it were so simple! No xtline isn't a command or
> and ado file or anything.
> > probably xtline will help?
> > I'm wondering if there's a way of analysing graphically the
> > of a panel regression (using xtreg or the xi: prefix). I'm
> doing some
> > FE estimations at the moment, and I want to draw some simple line
> > graphs over time of the residuals (and the fitted values).
> My data has
> > T>N, and I have concerns about non-stationarity, so this sort of
> > procedure is entirely necessary and/or useful.
> > However, I am only able to get legible graphs if I tell
> stata to plot
> > the residuals by each N. If I don't tell stata to treat each N
> > separately the graph is a right mess, because it draws a
> line for each
> > N, and connect the end of the first to the beginning of the second
> > etc. What I want is a set one set of residuals, not N sets, and one
> > set of yhats.
* For searches and help try: