Using same variable for weight and control

[James Park]

I would like to use the same variable for weight and control like:

sysuse auto
ivreghdfe price weight (length=trunk) [aw=weight], absorb(rep78)

But it generates error that says

Error: cannot use weight variables as dependent variable, regressor or IV in combination with -partial- option.

Is there any workaround?

[Carlo Lazzaro]

Caecilia;
my previous reply stated that what your seemingly are after is fesible in Stata; its methodological acceptability is a totally different topic.
The usual advice is skimming through the literature of your research field and see what others did in the past when presented with the same research topic.
As an aside, I'm not clear why not invoking -robust- standard error if you detected heteroskedastcity in rsidual distribution.

[Caecilia Lipowski]

Hi Carlo,

I want to include the same variable as control (it would introduce an omitted variable bias if I didn't) and as weight (to give more weight to certain observations and because of heteroscedasticity). Your answer suggests that this does not pose a problem. Could you please elaborate further or refer to some reference? In general, I was wondering if including the same variable as weight and as control makes any sense and more specifically, I was worried about consistency.

Thanks!

[Carlo Lazzaro]

James:
provided that it makes methodological sense, you can create a new variable that mirrors -weight-, but under a different name:

sysuse auto
clonevar Weight = weight
ivreghdfe price Weight (length=trunk) [aw=weight], absorb(rep78)