Location is one of the most important decisions for hotel chains and lodging firms. A hotel chain that can select good sites more accurately and quickly than its competition has a distinct competitive advantage.

Researchers S. E. Kimes (Cornell University) and J.

A. Fitzsimmons (University of Texas) studied the site selection process of La Quinta Motor Inns, a moderately priced hotel chain (Interfaces, Mar.-Apr. 1990).

Using data collected on 57 mature inns owned by La Quinta, the researchers built a regression model designed to predict the profitability for sites under construction.The least squares model is given below:

where y = operating margin (measured as a percentage)

- (profit + interest expenses + depreciation -

total revenue x1 = state population (in thousands) divided by the total number of inns in the state x2 = room rate ($) for the inn x3 = square root of the median income of the area (in

$ thousands)

x4 = number of college students within four miles of the inn All variables were "standardized" to have a mean of 0 and a standard deviation of 1. Interpret the j3 estimates of the model. Commcnt on the effect of each independent variable on operating margin, y. [Note: A profitable inn is defined as one with an operating margin of over 50% .]