Comparing Forecasting Models in Tourism

Abstract

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.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 54 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.