Dear Statalist users,

I have a panel dataset for 15 years and my dependent variable is binary in nature. Given the simultaneity issue, I want to apply an IV probit model.

following https://www.statalist.org/forums/for...ing-panel-data Prof Wooldridge advice on this forum, I use the cluster id option to generate valid inference.

However, using this option hinders me from using two-step estimation and hence I cannot obtain the first stage F statistics that I need to check the validity of my Instrument.

Is there an alternative way of doing IVprobit for a panel that reports first stage regressions diagnostic tests