New regression functionality: Hausman test for panel regression

It is now possible to perform model testing to check whether fixed or random effects estimation should be used in connection with panel regressions. This is done by using the command hausman.

The syntax and input follow the same logic as the various regression commands: The dependent variable is used as the first input, followed by the independent ones.

Example:

regress-panel salary age highedu wealth oslo
hausman salary age highedu wealth oslo

The result of the execution is a standard regression output for fixed and random effect estimation respectively. In addition, the difference between the coefficients from the alternative estimates is also shown, as well as an aggregate measure that indicates which variant is best to use for the current data set: P-value based on chi-square diagnostics.

P-values < 0.05 indicate that there are systematic differences in the coefficient estimates and that fixed effect modeling fits the data best. P-values above this limit indicate the opposite (that random effect modeling should be used).

Click here for example on panel regression using hausman test.

For more details, we recommend using the help command: help hausman

The new command regress-panel-predict can also be used as an aid tool for model diagnostics (click here).

We work continuously to improve microdata.no so that most analysis needs can be met. Thus, more and more analysis options will be introduced. We are based on input from researchers, and have compiled a list of new tools that we work through. Do you have specific suggestions for new functionality? Feel free to contact us by email: support@microdata.no