Kleibergen-Paap rk Wald F statistic output using iveghdfe

[Lorien Nair]

Hello everyone,

I am running a model using IV methods and wanted to report first stage F-stats, and since I do not make the i.i.d. assumption, I cannot use the Cragg-Donald Wald F statistic, and use Kleibergen-Paap rk Wald F Statistic. I know that the critical values for this are between 5.53 and 16.38. Thus is my understanding right if I infer from the results below that I have a weak instrument problem?

Weak identification test (Cragg-Donald Wald F statistic):        2.760
(Kleibergen-Paap rk Wald F statistic):        2.859
Stock-Yogo weak ID test critical values: 10% maximal IV size        19.93
15% maximal IV size        11.59
20% maximal IV size        8.75
25% maximal IV size        7.25
Source: Stock-Yogo (2005).  Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d.    errors.

Thanks!

[Alessandro Sovera]

That is correct.