I am wanting to use bsweights to resample the weights before a bootstrap a weighted regression. I do not a have a stratum to reference for the arguments in the bsweights command. I still need to provide a value for n() even though I let Stata know that my data is not in survey format. Stata will not allow anything but a positive value for n() for my data. I am nervous about choosing a value for n() since Stata creates a fake stratum for data not in survey format. Is it possible to choose a 'wrong' value for n() in this context? My Stata feedback is attached.