08-11-2014 05:31 PM
I'm running a VAR on monthly data, and I have two dummy variables I've included to eliminate outliars on 01may2010 and 01jun2010, (i.e. they are one on those dates and zero otherwise)
When I run a VARX model with two autoregresive lags and both dummy variables included I get the error message
"The model is not full rank when the forecasts of regressors are computed. The VARMAX
procedure stopped processing further steps."
Which is also a problem I had previously when I tried to include a trend term manually instead of through the "trend=linear".
My question is: is there any option for eliminating these outliers besides dummy variables, or is there a way to stop Varmax from computing the forecasts of regressors?