Dear Statalist,

I have a cross-sectional survey data on companies that was collected over several periods of time ("waves"). Each company in the survey is assigned with a certain weight to indicate its importance in the economy based on how many people the company employs. Moreover, the companies are selected randomly from a business database and stratified by a country, enterprise size class and economic activity. Each company appears in the survey only once and the weights are therefore unique. My independent variables are categorical, either nominal or ordinal.
The dependant variable is dichotomous and I've decided to go with a logit model. I'd like to use fixed effects as well as clustering at the country level but at the same time make use of the weights that were assigned to each individual firm. I'm wondering what type of weights would you recommend me using? My guess would be pweight and my sample code goes like this:

Code:
logit limited post##influenced ib3.var1 io9.var1 var2 ib2.var3 (...) i.wave i.country##i.industry [pweight=weight], vce (cluster country)
I haven't worked with weights before and would really appreciate your advice on this! Thank you very much!