Tourism Demand Forecasting - Other Methods
Methodology  -  Lake State Examples - Other Examples         
Back to Table of Contents                                                                                                   


Provencher, B. and R. C. Bishop. 2004. "Does Accounting for Preference Heterogeneity Improve the Forecasting of a Random Utility Model? A Case Study." Journal of Environmental Economics and Management 48:793-810.

This paper compares the recreation demand forecasting performance of a number of static random utility models: a logit model; two random parameters logit (RPL) models; and, a latent class (LCL) model. No model is clearly superior to the others and, surprisingly, by some measures a standard logit model does better in out-of-sample forecasts than models designed to capture angler heterogeneity. The results illustrate that with heavily parameterized econometric models and a choice model that is misspecified, the addition of parameters to denote the heterogeneity of preferences will "absorb" specification errors and thus possibly generate models inferior to those of a simpler model.

Chen, R. J., P. Bloomfield and J. S. Fu. 2003. "An Evaluation of Alternative Forecasting Methods to Recreation Visitation." Journal of Leisure Research 35(4):441-454.

This study examines the advantages and disadvantages of basic, intermediate and advanced methods for visitor use forecasting were seasonality and limited data are characteristics of the estimation problem. The monthly use rates at the Milwaukee County Zoo are used to illustrate the seasonal time series techniques. Forecasting methods evaluated include: two naive techniques, a single moving average (SMA) with the classical decomposition procedure, single exponential smoothing (SES), double exponential smoothing (DES, Winter's, and the seasonal autoregressive integrated moving average (SARIMA). SARIMA and SMA are found to the most appropriate methods in this case study. A useful comparative table is included listing the advantages and disadvantages of each method for predicting seasonal visitor patterns depending on the quality and characteristics of the data available for analysis.

Phaneuf, D. J. 1999. "A Dual Approach to Modeling Corner Solutions in Recreation Demand." Journal of Environmental Economics and Management 37:85-105.

The dual approach of Lee and Pitt to estimating demand systems for which individuals often choose not to consume one or more of the available goods provides a utility-consistent framework for estimating preferences of visits to recreation sites. Due to the complexity of the model, however, this approach has not been applied to in the recreation demand literature. This paper provides a first application of this method to demand for fishing in Wisconsin and welfare analysis is conducted for four policy scenarios.

Provencher, B. and R. C. Bishop. 1997. "An Estimable Dynamic Model of Recreation Behavior with an Application to Great Lakes Angling." Journal of Environmental Economics and Management 33:107-127.

Travel cost method is widely applied to estimate the economic benefit of nonmarket resources for site-specific recreational activities. This paper develops a dynamic structural model of the decision to visit a recreation site. Compared to the typical static model approach to this problem, a dynamic model allows the analyst to develop a decision problem that looks more like "the real thing". For illustration, the model is applied to the decisions of of fishing club members on the Wisconsin shore of Lake Michigan. The authors conclude that due to the challenges of obtaining appropriate data and some of the limiting assumptions of the model, that this type of model is likely appropriate only in certain circumstances. In many cases the static model will likely yield welfare estimates similar to the dynamic model with much less cost and effort. The relative accuracy of each modeling technique needs more empirical investigation.

Lake States Examples:

Somersan, A. and R. Shaver. 1987. Market Study and Economic Impact Assessment for Sheboygan Harbor Marina Slips. Madison, WI: Recreation Resources Center, University of Wisconsin-Extension.

The City of Sheboygan in considering the construction of a new marina needed to assess its viability. This study assessed the demand and supply of marina slips in the area and estimated the economic impact of a marina in Sheboygan. Using secondary data, the authors concluded that boating was a growing activity and the supply of slips was inadequate. Total spending with a 200-slip marina in place would amount to about $3 million with an estimated 64 jobs generated. The authors warn that the costs of the marina must be carefully looked at and planned for.

Drewiske, D. 1984. Economic Impact Potential for the Racine Harbor Development Project. Madison, WI: Recreation Resources Center, University of Wisconsin-Extension.

This study analyzed the planned improvements for the Racine Harbor. A demand analysis was conducted and found that the planned marina rehabilitation will begin to fill demand. The marina was expected to have a positive economic impact on employment, income, and the public sector. Total economic impact was estimated at close to $20 million. The number of full-time equivalent jobs was estimated to be about 400.

Other Examples:

Loomis, J. B. 1995. "Four Models for Determining Environmental Quality Effects on Recreational Demand and Regional Economies." Ecological Economics 12(1995):55-65.

This paper addresses the paucity of research which links recreational demand modeling with regional economic analysis modeling. The choice to participate in a recreational activity at a particular site are based on four related recreation choices: 1) decision to participate in a given recreation activity; 2) decision about which of the available sites to visit; 3) decision about the frequency of trips to take to a given site; and 4) decision about length of stay at the recreation site. Each of these four recreational choice decisions is related to and influenced by environmental quality and site facilities. When modeling the economic impact of the improvement (or degradation) of environmental quality or site facilities it is important to consider all four factors or risk underestimating the economic impact of the change.

Brown, T. L. and N. A. Connelly. 1994. "Predicting Demand for Big Game and Small Game Hunting License: The New York Experience." Wildlife Society Bulletin 22:172-178.

The paper presents of the results of a study using ordinary least squares regression to model big game and small game license sales. A relatively limited and inexpensive database that included license sales, license cost, demographic variables and available resources data for New York for 1962-1991 was used in the modeling process. The independent variables used in the analysis included: 1) license cost and ability to pay (e.g. income); 2) size of the general population to which hunters belong; 3) degree of urbanization; 4) access to the resource; and 5) supply of games species and perceived probability of harvest success. Time series regression analysis of license sales offers a seldom-used opportunity for state wildlife agencies to improve their understanding of the demand for small and big game hunting at the state level.

Tay, R. S. and P. S. McCarthy. 1994. "Benefits of Improved Water Quality: A Discrete Choice Analysis of Freshwater Recreational Demands." Environment and Planning 26(10):1625-1638.

Discrete choice methodologies are often used estimate multiple-sites recreational demands and evaluate the welfare effects of alternative environmental policies aimed at water quality improvements. This study uses 1985 data on Indiana anglers to estimate a multinomial logit model of destination choice and compute the benefits of alternative water quality improvements. In general, the results indicate that anglers are reasonably sensitive to changes in water quality. The per-trip welfare gains from a 1% reduction in various pollutants range from 4.9 to 25.3 cents and a similar reduction in all-pollutants increases per-trip welfare by 64.5 cents.

Lieber, S. R., D. R. Fesenmaier and R. S. Bristow. 1989. "Recreation Expenditures and Opportunity Theory: The Case of Illinois." Journal of Leisure Research 21(2):106-123.

This paper investigates the factors that affect decision making of recreators. In particular the spatial context within which recreators make destination choices in considered. The relative effects of site characteristics are compared to the spatial context using a multiple regression model. Significantly, the results show that agglomerative effects and contextual effects of spatial structure are the principal factors influencing per person per day expenditure levels. The number of other facilities within 20 miles of the chosen destination had the most impact on per person per day expenditures. Facility development was shown to be the dominant predictive force in accounting for the level of expenditures.

Lindeborg, K. H. 1973. Evaluation of Regional Multipurpose Economic Benefits Resulting from a Water and Related Land Development. Research Technical Completion Report Project C-2195-IDA. Moscow, ID: Water Resources Research Institute, University of Idaho.

This study evaluates the economic impacts of water resources on seven major uses: irrigation, recreation, power, municipal and industrial use, flood control, water quality and fish and wildlife. To estimate the economic impact of recreation (pheasant hunting only) a simple demand model was constructed based on the concept of marginal utility. Three factors were considered: travel distance to recreation site, irrigated cropland as the sites of recreation and hunter success. The model predicted a small increase in total economic impact, but there were regional disparities. The author pointed out the results lacked reliability due to the data used, but indicated that the general approach had good value if better data were used and the analysis expanded to include other recreation activities.

Michalson, E. L. 1973. Recreational and Sociological Characteristics of Hunters and an Estimate of the Demand for Hunting in the Sawtooth Area of Idaho. Scenic Rivers Study Report No. 7. Moscow, ID: Water Resources Research Institute, University of Idaho.

This study surveyed hunters in the Sawtooth Area of Idaho. A linear regression demand model was constructed based on the variables of: miles traveled, cost per visitor day, education level, annual income and number of trips made by hunters in 1971. The model was used to calculate a demand curve and then estimate willingness to pay and total net value of the hunting resource in the Sawtooth Area.

Stevens, T. H. and R. J. Kalter. 1970. Technological Externalities, Outdoor Recreation, and the Regional Economic Impact of Cayuga Lake. A. E. Res. 317. Ithaca, NY: Department of Agricultural Economics, Cornell University Agricultural Experiment Station, New York State College of Agriculture.

This study quantified the economic impact of Cayuga Lake in New York State on the regional economy. A demand allocation model was developed based on nationwide survey of preferences for recreation use. Both willingness-to-pay and estimated direct and indirect expenditures were then calculated based the demand allocation model and results from other studies. The final results provided an estimate of the total economic impact of recreation on Cayuga Lake and provided a baseline for estimating how a proposed thermal plant might affect the regional economic impact of recreationists.

Back to Table of Contents