st: predict, option dynamic, confidence intervals

[Combes Stephanie <Stephanie.Combes@ensae.fr>]

Hello,

I am surprised that confidence intervals are not optional in predict, I
would like to predict 5 years ahead the end of my datas, so I use the
dynamic option, and I would like to get the corresponding confidence
interval (which should be widening), are them some code done for this?

For my prediction, I'm using an ARIMA modelisation, I don't really get the
help explication about yresiduals and residuals, yresiduals = y - yhat? What
is "residuals" then?

I would be very grateful if someone has an answer for me, thank you by
advance for your time and help,
With regards,

Stephanie Combes

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/