On the Relationship between Length of Stay and Total Trip Expenditures: A Case Study of Instrumental Variable (IV) Regression Analysis

Abstract

The relationship between length of stay (LOS) and total trip expenditures (TTE) has been scrutinized many times within a micro-econometric framework, usually by means of ordinary least squares (OLS) regression analysis. The author questions this practice because much evidence suggests that LOS is an ‘endogenous' independent variable. One of the basic assumptions of OLS regression is thus violated, and a new method — instrumental variable (IV) regression — is called for to produce a consistent, unbiased estimate of LOS. A non-technical case study on IV regression shows that the LOS—TTE relationship estimated by IV regression analysis is only about half the analogue OLS estimate. The study concludes with several important implications for the statistical modelling of micro-level tourism expenditures and for cross-sectional regression-based tourism studies in general.