01-21-2016 08:56 PM
SAS provides an example about how to fit a mixed logit model using PROC MDC. Below is the link of the example:
There is a table at the bottom of the webpage which shows the results of parameter estimates. Here are my questions:
(1) What is the meaning of the last colume in the table? When the value of the 'Approx Pr > |t|' is smaller than 0.05, does that mean 'it is NOT necessary to consider random effect for that variable (e.g., ttime)'?
(2) What kind of test is used here? the t-test?
(3) Where can I find more introduction about the fitting results?
Any comments or suggestions will be highly appreciated!