topic Re: Predicting outcome in Sample Selection Model using PROC QLIM in SAS Forecasting and Econometrics

Predicting outcome in Sample Selection Model using PROC QLIM

ZMX — Thu, 22 Jun 2023 20:56:01 GMT

I was looking at the example below on the SAS website and I was wondering how to use the results of this model to predict unconditional wages (ie. E(w) versus E(w|inlf=1).

Are the wages predicted using the second regression conditional wage (ie. E(wage|inlf=1) )?

proc qlim data=mroz;
      model inlf = nwifeinc educ exper expersq
                   age kidslt6 kidsge6 /discrete;
      model lwage = educ exper expersq / select(inlf=1);
   run;

Thanks for the help.

Re: Predicting outcome in Sample Selection Model using PROC QLIM

SASCom1 — Tue, 18 Jul 2023 21:43:12 GMT

Hello @ZMX ,

You can obtain unconditional expected value and conditional expected values using the EXPECTED and CONDITOINAL options in the OUTPUT statement directly in PROC QLIM.

For the sample selection model, the PREDICTED value is equal to the EXPECTED value, which is also equal to the XBETA value for the continuous dependent variable in the main equation(lwage in the example). You can verify this by specifying the OUTPUT statement like below:

output out = outdata expected xbeta predicted conditional mills ;

and look at the outdata data set.

The conditional expected value differs from the expected/predicted/xbeta values by adding to it another term, rho*sigma_lwage*Mills_inlf, in other words,

Predicted_lwage = Expected_lwage = xbeta_lwage ;

Conditional_lwage = xbeta_lwage + rho*sigma_lwage*Mills_inlf ;

I hope this helps.

Re: Predicting outcome in Sample Selection Model using PROC QLIM

SASCom1 — Tue, 18 Jul 2023 21:45:19 GMT

@ZMX ,

Sorry, in my previous post, 'CONDITOINAL' is a typo, it should be CONDITIONAL 🙂

Re: Predicting outcome in Sample Selection Model using PROC QLIM

ZMX — Thu, 16 Nov 2023 03:25:58 GMT

It did help. Thank you for the response. 🙂