Abstract

A model of retail activity which focuses on firm location, firm size, and intensity of capital use is developed. This model has discrete and continuous dependent variables. Owing to its nonlinearity, a number of approximation techniques to facilitate parameter estimation are explored. The two promising approaches require making floor space a discrete variable. The first method places restrictions on the structure of unobserved variables in the discrete choice of floor space and location. The second approach uses the multinomial probit model of discrete choice. The latter model is more general than the former but will require the modification of existing probit estimation procedures.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call