forecasting production with Stata 12

[Leah Lee]

Hello.
I'm Leah Lee from Korea.
I would like to use the STATA 12 to estimate the next 5 years(monthly) of production from the time-series data.
My monthly data is from January 2009 to December 2016.


However, as I mentioned above, I would like to ask you a question as it is difficult to predict the output over the next five years.
Is there any way to forecast(predict) with STATA 12? Because I only have this version.

Thank you.

[Ariel Karlinsky]

I don't have much experience with time series models, but can't you just tsappend to add new periods and then predict with a dynamic condition to allow for prediction using predicted values of variables where these are missing?

code example:

webuse air2, clear
tsset
generate lnair = ln(air)
arima lnair, arima(0,1,1) sarima(0,1,1,12) noconstant
tsappend, add(56)
predict lnair_hat, y dynamic(144)
tsline lnair lnair_hat, xline(144)

[Leah Lee]

Hello. Mr.Ariel.
First of all, thank you for your reply.
I'll tell you how I tried.

Code :
Tsappend, add (36)
QuI arima product, ar (1) nolog
predict y if time>tm(2016m12)
I used ln_product and D.ln_product to replace the product variables, and got the following value.

I am curious about the estimated value from January 2017 to December 2019, but I got the same price.
Please let me know if I miss anything.
Thank you.

[Ariel Karlinsky]

why predict only after 2016m12? predict for the whole time, just use the dynamic option of predict to allow the prediction to rely on predicted values and not just on originals (see help arima post-estimation for details).