Hi,
I am trying to perform a stepwise regression with a lot of independent variables (790), and with a 5% p-value significance level for all the variables. I am using Stata 16.
First I do:
reg y q1 q2 q2 q3 H I J K L M N O...(all the other Xs)
then I do stepwise, pr(0.05) : regress y (q1 q2 q3 q4 H I J K L M N O.. (all the other Xs))
But I got this error:
begin with full model
last test not found
r(302);
Can someone kindly help me? Let me know if you need more information. I didn't put all the other variable names as they are 790, but it is not a variable problem as the first reg command goes through perfectly (with al lot of variables dropped because of estimability).
Carlo
Related Posts with Stepwise Regression Issue
Monte Carlo--How to see B variabilityHi everyone, I created a monte carlo simulation. I am trying to see attentuation bias which I see, …
New Stata User: Need help with effect size question and one other using svy tabulateI have a survey table with two variables. After output it shows 2x3 table and measures off associati…
Using -foreach- to count for string variablesDear Statalists, I have a dataset with 28 string variables named"ox*tc", and the * represent to num…
Correlation between binary and continuous variablesHello everyone! I have a bunch of binary variables, such as GENDER, and some contiuous variables li…
Monte Carlo FGLSI have the following DGP: "Data generating process (DGP): Assume the true data generating process is…
Subscribe to:
Post Comments (Atom)
0 Response to Stepwise Regression Issue
Post a Comment