How can I deal with unequal sample sizes in STATA?

#1
I am currently writing my thesis about ESG scores and the influence on firm value. I want to see whether there are differences between advanced and emerging markets. However, the sample I use is totally not equally distributed. As the data for firms in emerging markets is harder to retrieve, I 'only' have 3,980 observations vs 15,983 observations for advanced markets. I use the time period 2013-2017 and the observations here are also not equally distributed. I have read about here about the problem related to the type II error rates caused by the highly unequal n. The writer suggested to do "compromise power analysis".However, I am not really advanced with statistics, can someone give me direction how to do this in STATA? I tried to google it, but it still seems ambigious to me.

Thank you in advance!
 
#3
@GretaGarbo

ESG = Environmental, Social and Governance

I want to see how ESG scores influence the stock price by controlling for the market (advanced or emerging)

Is it more clear now?

Thanks for helping GretaGarbo
 
#4
ESG = Environmental, Social and Governance
What kind of numbers does that give?

So the ESG scores will explain the stock value? So that you have a model like:

Stock_value = a + b* ESG + other_things

Is not every stock valued at completely different scales? How can you compare them?