Dear Statalisters,

I am using Stata ver 14, and the built in command aidsills written by Lecocq and Robin (2015).
I am estimating demand for 4 different "brands".
I am looking to apply the instrumental variable technique of Hausman, Leonard and Zona (1994), which they apply on their estimation of demand for beer brands, that is, using neighboring city's prices as price instruments. However, this seems to hinge on the fact that you include brand- and city-fixed effects.
However, when I try to include this in my demand system with aidsills, i retrieve an error message stating
"estimates repost: matrix has missing values
r(504);"
Does anyone know if it is possible to include brand- and city-fixed effects with aidsills?

References:
Sébastien Lecocq, Robin, Jean-Marc: "Estimating almost-ideal demand systems with endogenous regressors." The Stata Journal Volume 15 Number 2: pp. 554-573
https://www.stata-journal.com/articl...article=st0393

Hausman, Jerry, Gregory Leonard, and J. Douglas Zona. "Competitive analysis with differenciated products." Annales d'Economie et de Statistique (1994): 159-180.

Best regards,
Hanna Lindström