panel regression model

[Lea June]

I am a management student, currently writing my master thesis. I would be grateful for any help!

The purpose of my research is to analyse if FDI is a channel of knowledge spillovers in the region of the Western Balkans

The dependent variable is productivity based on the available data for a period of 10 years and 6 countries/

The other variables are:

• FDI as a % of GDP
• Inflation
• Competition, GDP per capita, and trade openness as a proxy for the absorptive capacity
• Level of corruption and political stability as a proxy for the institutional quality

My question:

To check whether I should use a fixed-effects or random-effects model, I did the Hausman test, but the output does not seem right.

From the p-value, I doo not reject the null hypothesis so in theory I should go for the RE but it appears that "V_b-V_B is not positive definite".

I also tried by adding "hausman fixed random, sigmamore", but that does not change anything to my results.

What would you recommend me to do? Does that simply mean I will have to stick to the random-effects model? And I should simply ignore the "V_b-V_B is not positive definite" ?


Please find a screenshot of the test attached.

[Carlo Lazzaro]

Lea:
what does -xtest0- tell you?

[Lea June]

Hi thank you for your answer, when I try to run this command it gives me "last estimates not xtreg, re" even though I test it for a fixed effect model I don't understand why. (I use Stata SE 18,5)
Best,
Lea

[Lea June]

.

[Carlo Lazzaro]

Lea:
-xtest0- works after -xtreg,re- only .

[Lea June]

this is what it says

[Carlo Lazzaro]

Lea:
-xtest0- outcome tells you that the use of -re- estimator is not appropriate in your case.