2SLS and over and under identification

[MARIA CHRISTOFIDOU]

I get the below results and I need help to interpret them. Is my IV a good one?
Are my results good?

The Cragg-Donald Wald F is 385,9 and teh Stock-Yogo at 10% 16,38 an at 15% 8,96.

I would also be grateful if you can give references to a book or publication/web sites that explains those statistics,
Much appreciate any help.

thank you

[Andrew Musau]

See the documentation for ivreg2 from SSC under "Tests of under and weak identification", where the Cragg-Donald Wald F-statistic and the Stock-Yogo critical values are discussed. Literature references are also included.

Are my results good?

To answer that, one needs knowledge of your data, instruments, research question, and the existing literature. If your question is simply about how to interpret those values: the null hypothesis of this test is that the instruments are weak—i.e., only weakly correlated with the endogenous regressor. If the Cragg-Donald F-statistic exceeds the Stock-Yogo critical value, you reject the null at that level.

In your case, 385.9 ≫ 16.38 and 8.96. So, you reject the null of weak instruments at both the 10% and 15% thresholds. If everything else in your model holds, then yes—your results look good.