Pooled panel regression with double clustered standard errors and industry fixed effects

[Karoline Karlsson]

Hi all,

I am trying to replicate a study on ESG disagreement and stock returns. I am fairly new at Stata and could use some help for this.
I have monthly stock returns as my dependent variable, and I have the standard deviation of ratings(STDEV) and the Fama French 5 factors as my explanatory variables.

I need to do a Pooled panel regression with double clustered standard errors (month and firm) - can anyone tell me the commands in Stata for doing this? And how do I incorporate industry-month fixed effects? Do I need a variable for the firms industry to do this?

Thank you all in advance.

Kind regards,
Karoline

[Fei Wang]

You may install and refer to reghdfe (from SSC) which helps easily control for fixed effects and implement multi-way clustering.

[Karoline Karlsson]

Hi Fei,

Thank you for your reply.

Could you write an example of how to use that command?

Thank you.

[Jared Greathouse]

You can just look at the help file. But an example they provide is

sysuse auto, clear




cls



reghdfe price weight length, absorb(rep78) vce(cluster rep78 foreign)