ScholarLens Back to search
2015•Tourism EconomicsRequires access

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

Christer Thrane

Open publisher page 23 citations

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.

About this research paper

What this paper is about

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.

Why it matters

OpenAlex reports 23 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available 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.

Key concepts: Ordinary least squares, Instrumental variable, Regression analysis, Econometrics, Regression, Variables, Statistics, Linear regression

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
On the Relationship between Length of Stay and Total Trip Expenditures: A Case Study of Instrumental Variable (IV) Regression Analysis — Research Paper | ScholarLens