Fcast Compute

[Manuel Gloria]

Good afternoon,

I am trying to use Fcast Compute after having estimated a VAR model with 3 variables and 4 lags but Stata (although correctly creating 12 new variables) is populating only the three SE variables, leaving blank the forecasted variables, upper bound and lower bound.

Am I missing something?

These below the codes I used

var dinf dfun dsav, lags(1/4)
fcast compute F_, step(10)
fcast graph F_dinf F_dfun F_dsav

Thanks,
Manuel

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex.

I don't do var so I can't help much. I'd wonder if there isn't something odd about your var estimates or whether fcast isn't just supposed to do standard errors on the predicted dv's (the three listed). I'd also see if the replace option helped. Sometimes it helps to replicate the results in the documentation and then modify to your variables.