This Article Full Text (PDF) All Versions of this Article: 1096348007309566v1 32/1/3    most recent References Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Add to Saved Citations Download to citation manager Reprints and Permissions Add to My Marked Citations Google Scholar Articles by Chen, R. J. C. Articles by Cubbage, F. W. Social Bookmarking                   What's this?

Comparing Forecasting Models in Tourism

University of Tennessee, rchen{at}utk.edu

Peter Bloomfield

North Carolina State University, bloomfld{at}stat.ncsu.edu

Frederick W. Cubbage

North Carolina State University, fred_cubbage{at}ncsu.edu

This study uses three major U.S. national parks as applications of statistically selecting appropriate methods to forecast attendance. Forecasting methods assessed include Naïve 1, Naïve 2, single moving average (SMA), single exponential smoothing (SES), Brown's, Holt's, autoregressive integrated moving average (ARIMA), derived time series cross-section regression (TSCSREG), and time series analysis with explanatory variable models. The mean absolute percentage error (MAPE) is used to measure the accuracy of forecasting methods. Based on the MAPE values, SMA produces the most accurate forecasting, followed closely by ARIMA, Brown's, and Naïve 1 models. Holt's and TSCSREG models produce the next most accurate forecasting, followed by SES, time series analysis with explanatory variable model, and Naïve 2. Methods used in this article are readily transferable to other hospitality and tourism data sets with annual visitation figures. Merits and limits of the proposed forecasting methods are discussed.

Key Words: mean absolute percentage error • forecasting methods

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?