Hi all,
I am relatively new to Stata and was looking for some direction on how to use the postestimation tools for ologit. Specifically, I am trying to take the estimates of an ordered logit regression in dataset #1, and predict the outcomes for that categorical variable in dataset #2 .
So far, I have been able to use the 'estimates store' and 'estimates restore' commands, along with 'predict' to get the predicted probabilities of each outcome for one categorical variable (which has six possible outcomes). Now, I am hoping to predict a particular value (e.g, 1-6) for that categorical variable using those probabilities. Is there some sort of random number function that would allow me to manually assign probabilities of particular outcomes? Or is there something that I am missing in the ologit postestimation documentation?
See some example code below:
I am relatively new to Stata and was looking for some direction on how to use the postestimation tools for ologit. Specifically, I am trying to take the estimates of an ordered logit regression in dataset #1, and predict the outcomes for that categorical variable in dataset #2 .
So far, I have been able to use the 'estimates store' and 'estimates restore' commands, along with 'predict' to get the predicted probabilities of each outcome for one categorical variable (which has six possible outcomes). Now, I am hoping to predict a particular value (e.g, 1-6) for that categorical variable using those probabilities. Is there some sort of random number function that would allow me to manually assign probabilities of particular outcomes? Or is there something that I am missing in the ologit postestimation documentation?
See some example code below:
Code:
use "dataset1.dta" ologit cat_var predictor1 predictor2 predictor3 predictor4 predictor5 estimates store m1 use "dataset2.dta", clear estimates restore m1 predict cat_var1 cat_var2 cat_var3 cat_var4 cat_var5 cat_var6
