Power calculations with IV

[Vishal Ahuja]

Hello (from a new member),

Can someone guide me on how to do power calculations in stata to show that the IV is capable of ruling out large effect sizes? I have a binary endogenous regressor (X), a binary outcome (Y) and an IV that is continuous (Z). My logit, OLS, and IV estimates are consistent in terms of direction and significance.

Thanks.

[Phil Bromiley]

Welcome to Statalist and thank you for providing data. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. We don't know what you really ran so interpreting your last sentence is very difficult.

I'm not sure exactly what you mean by power analysis in the context of effect sizes. Most power analyses I know of address whether the sample is large enough to reliably find a significant effect if the variable explains a given amount of the variance in the dv. I suggest you start a new threat with a much clearer explanation of your problem.

By the way, please use Stata not stata.