R2 for xtnbreg model

[Jana Schue]

Hello everyone,

i am currently analyzing a count variable with the -xtnbreg- command.
This command does not show me the R2 of my model, but it would be quite helpful for the interpretation of my analysis.
Is there a chance to get the R2 for my count variable by unsing a specific command?


Best regards
Jana

[Carlo Lazzaro]

Jana:
as -xtnbreg- returns -e(chi2)-, R2 is not reported (and it would not be useful for the interpretation of your analysis, as you're not dealing with a linear regression model).

[Jana Schue]

Okay thanks for clarification Carlo.

The results of my analysis are:
Wald chi2(15)=761,75
Prob > chi2 = 0.0000

Could you give me an advice how to interpret them or any link where I can learn it.
The staitisical education of my univerity was sadly pretty poor.

Kind regards,
Jana

[Carlo Lazzaro]

Jana:
as per all the regression models that uses MLE, -chi2- tests the joint statistical significance of the predictors plugged in the right-hand side of your regression equation vs

an empty model (i.e., a model with no predictors).
Your results show that your model with predictors is more informative than its empty counterpart.

A possible link (that includes references) is https://stats.idre.ucla.edu/stata/da...ic-regression/.

[Jana Schue]

Thank you so much!
Carlo, you helped me a lot

Jana