Microsoft Word 2007 ichrie conference Proceedings Final-Final 06-06-07. doc

METHODOLOGY 
Visitor Spending Profiles
This study used the spending profile compiled by Chen in 1996 and inflated the 1995-1996 money figure to 2007 
money figure. Readers are referred to Chen (1996) for the detailed data collection. This study used an input-output 
model to produce quantitative estimates, sector-by-sector of the economic impact of tourism and related industries 
of the regional economy along the Blue Ridge Parkway. 
Forecasting Methods 
Short- and medium-term forecasts can provide valuable information for pricing, seasonal employment, and 
short-term budgeting (Chen, Bloomfield, and Fu, 2003). With respect to developing long-run strategic plans and 
sustain the existing resources of a regional destination, forecasts can determine future infrastructure needs, 
community development, new attractions, and future employment (Chen, 2006b). The suggested forecasting 
methods in this study may include (1) Naïve models, (2) Holt Winter’s seasonal double exponential smoothing 
model, and (3) Autoregressive Integrated Moving Average (ARIMA) model. Forecast accuracy may be assessed 
using mean absolute percentage error (MAPE) and root mean square percentage error (RMSPE). According to 
previous forecasting studies (Chen, Bloomfield, & Fu, 2003; Chen, 2006a), ARIMA was the best choice. This study 
mainly employed the ARIMA forecasting model (Table 1) to forecast the Blue Ridge Parkway’s visitation figures.
For more information about the calculation and limitations of the ARIMA method, readers are referred to Box, 
Jenkins, and Reinsel (1994).