Degree Type

Dissertation

Date of Award

2018

Degree Name

Doctor of Philosophy

Department

Apparel, Events and Hospitality Management

Major

Hospitality Management

First Advisor

Robert Bosselman

Abstract

This research study examined impact of oil prices on RevPAR. Most are familiar with the impact of higher oil/energy prices on broader travel trends as it generally causes less hospitality spending. Few have looked at the operating impact of oil prices on hotel and travel consumption in markets dependent on the commerce this commodity brings.

Quantitative statistical methods were employed utilizing time series analysis aimed at testing a proposed model of variation in monthly RevPAR within shale-producing and non-shale producing regions between 1990 and 2016, as well as during periods of rising and declining oil prices. Predictors of RevPAR in the proposed model included: Oil Prices, Room Supply, Unemployment rate, and Personal Income.

There was considerable autocorrelation in the variables, and substantial evidence of autocorrelation in the residuals for the models without adjusting for the auto correlation the initial analysis results in the relatively high R-square between RevPAR and the independent variables. However, when residualized variables were included that controlled for the autoregressive, integrated and moving average components of these variables, the amount of variance explained in RevPAR dropped considerably.

After reducing the autocorrelation, examining different time periods, and markets that are closely aligned with oil production or the broader U.S. there was not a statistically significant causal relationship between oil prices and RevPAR. Furthermore, the findings implied that it does not seem to matter whether the markets being studied are oil producing or not as the relationships are not significant. Hospitality leaders may have inadvertently “blamed” weakness in overall RevPAR and RevPAR in shale markets on lower oil prices during 2015 and 2016, while this analysis was less conclusive of this relationship. Industry experts sometimes publish high R squares to imply greater certainty of the relationship between independent variables and RevPAR, but these equations need to be tested for autocorrelation.

Key words: RevPAR, shale oil, autocorrelation and Seasonal ARIMA

Copyright Owner

Steven Eric Kent

Language

en

File Format

application/pdf

File Size

100 pages

Share

COinS