Hausmann Test: the rank of the differenced variance matrix (2) does not equal the number of coefficients being tested (3)

[Jasmin Music]

Hi,

I perfromed a Hausman test to see whether I should use a dfixed or random effects model. The result show that I should use fixed effects model. However I do not understand the NOTE under it.
Can someone just please tell me, if I can still argue why I went for the fixed effects model, based on the Dubrin-Wu-Hausman test?

Thanks in advance and best regards

Jasmina

[George Ford]

https://www.stata.com/statalist/archive/2012-01/msg00092.html

might want to scale the municipalities variable so its coefficient is closer to 1.

[Jasmin Music]

Thanks a lot. How can I do that in my specific case (scale municipalities) ?

I have already run a fixed effects model. Because one profesor suggested it and the hausman test showed so, but I want to be sure the test is performed correctly and that I can rely on the hausman test.

[Carlo Lazzaro]

Jasmin:
as an aside to George's helpful advice, you may want to take a look at https://www.statalist.org/forums/for...s-being-tested

[Jasmin Music]

Great Carlo, thanks! This helped me a lot. I checked that issue, and in my case also no observations and groups are dropped, so I will continue with fixed effects model, since my p-Value is lower than the significance level.

[George Ford]

I'm not sure what the municipalities variable is, but if continuous to rescale you just multiply/divide the variable to shift the decimal on the estimated coefficient.

[Andrew Musau]

Can someone just please tell me, if I can still argue why I went for the fixed effects model, based on the Dubrin-Wu-Hausman test?

You do not need a Hausman test to justify using the fixed effects model. The fixed effects estimator is consistent whether or not the random effects assumption holds. You test only if you are considering using the random effects estimator. See https://www.stata.com/support/faqs/s...effects-model/

[Jasmin Music]

Hi Andrew, can you please again send me the appropriate page, because tghis page was not found when entering the link?

[Andrew Musau]

Click on the link (instead of copying and pasting).