Stock-Yogo weak ID test critical values by ivreg2h command

[Aleksandra Anic]

Dear STATA users,

I use ivreg2h user written command by Baum, CF, Schaffer, ME, 2012. ivreg2h: Stata module to perform instrumental variables estimation using heteroskedasticity-based instruments. The command is based on Lewbel 2012 and 2016 approach. The command calculates among many other tests also Kleibergen-Paap rk Wald F statistic for weak identification. I am confused about what critical values should I use to make conclusions, because there are two different sets of critical values.
Stock-Yogo weak ID test critical values based on maximal IV relative bias and maximal IV size. Should I compare test statistic with "relative bias" or "size"?

The critical values are listed as follws
Stock-Yogo weak ID test critical values: 5% maximal IV relative bias 21.28
10% maximal IV relative bias 11.50
20% maximal IV relative bias 6.39
30% maximal IV relative bias 4.59
10% maximal IV size 52.77
15% maximal IV size 27.99
20% maximal IV size 19.51
25% maximal IV size 15.19

Thank you very much on your answer.

Best wishes,
Aleksandra