The Implications of Model Specification for Welfare Estimation in Nested Logit Models

Kling, Catherine L.; Thomson, Cynthia J.

doi:10.2307/1243782

Cited by 80 publications

(37 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At best, in these cases travel costs can be bounded by considering the nearest and furthest access points along the river segment. Similar problems emerge in the context of beach (e.g., [13]) and wetland (e.g., [20]) recreation, or, more generally, any large geographic regions.…”

Section: Introductionmentioning

confidence: 76%

“…An error component structure similar to specification 1 of the nested logit model in the previous section would replace (13) with…”

Section: Specification 1: Trip Nestmentioning

confidence: 99%

“…The issue with these ad hoc approaches is that they implicitly make assumptions regarding the role of travel costs (the marginal utility of income) in the individual's decision making that are inherently inconsistent with the broader models used to represent the choice among river segments. For example, using the nearest access point along a river segment to compute travel cost implicitly assumes that travel cost is the determining factor in choosing where along a river segment to recreate (essentially assuming that the marginal utility of income is infinite), whereas in the broader models of the choice among river segments (say a RUM model) travel cost is but one of the factors in site selection (implying a finite marginal utility of income) (e.g., [13] and [20]). …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Modeling Recreation Demand When the Access Point Is Unknown

Herriges

Kling

2016

American J Agri Economics

Self Cite

View full text Add to dashboard Cite

The task of modeling the recreation demand for geographically large sites, such as rivers and beaches or large parks with multiple entrances, is often challenged by incomplete information regarding the access point used by the individual. Traditionally, analysts have relied upon convenient approximations, defining travel time and travel distances on the basis of the midpoint of a river or beach segment or on the basis of the nearest access point to the site for each individual. In this paper, we instead treat the problem as one of aggregation, drawing upon and generalizing results from the aggregation literature. The resulting model yields a consistent framework for incorporating information on site characteristics and travel costs gathered at a finer level than that used to obtain trip counts. We use a series of Monte Carlo experiments to illustrate the performance of the traditional mid-point and nearest access point approximations. Our results suggest that, while the nearest access point approach provides a relatively good approximation to underlying preferences for a wide range of parameter specifications, use of the midpoint approach to calculating travel cost can lead to significant bias in the travel cost parameter and corresponding welfare calculations. Finally, we use our approach in modeling recreation demand for the major river systems in Iowa using data from the 2009 Iowa Rivers and River Corridors Survey. Disciplines Agricultural and Resource Economics | Economics | Natural Resource Economics AbstractThe task of modeling the recreation demand for geographically large sites, such as rivers and beaches or large parks with multiple entrances, is often challenged by incomplete information regarding the access point used by the individual. Traditionally, analysts have relied upon convenient approximations, defining travel time and travel distances on the basis of the midpoint of a river or beach segment or on the basis of the nearest access point to the site for each individual. In this paper, we instead treat the problem as one of aggregation, drawing upon and generalizing results from the aggregation literature. The resulting model yields a consistent framework for incorporating information on site characteristics and travel costs gathered at a finer level than that used to obtain trip counts. We use a series of Monte Carlo experiments to illustrate the performance of the traditional mid-point and nearest access point approximations. Our results suggest that, while the nearest access point approach provides a relatively good approximation to underlying preferences for a wide range of parameter specifications, use of the midpoint approach to calculating travel cost can lead to significant bias in the travel cost parameter and corresponding welfare calculations. Finally, we use our approach in modeling recreation demand for the major river systems in Iowa using data from the

show abstract

Section: Introductionmentioning

confidence: 76%

“…An error component structure similar to specification 1 of the nested logit model in the previous section would replace (13) with…”

Section: Specification 1: Trip Nestmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling Recreation Demand When the Access Point Is Unknown

Herriges

Kling

2016

American J Agri Economics

Self Cite

View full text Add to dashboard Cite

show abstract

“…A complete description of the data can be found in Thomson and Crooke (1991) and Kling and Thomson (1996). Random telephone interviews were conducted in Southern California to identify recreational anglers, who were then requested to complete a follow-up mail questionnaire.…”

Section: Datamentioning

confidence: 99%

Nonlinear Income Effects in Random Utility Models

Herriges

Kling

1999

Review of Economics and Statistics

Self Cite

129

View full text Add to dashboard Cite

Random utility models (RUMs) are used in the literature to model consumer choices from among a discrete set of alternatives, and they typically impose a constant marginal utility of income on individual preferences. This assumption is driven partially by the difficulty of constructing welfare estimates in models with nonlinear income effects. Recently, McFadden (1995) developed an algorithm for computing these welfare impacts using a Monte Carlo Markov chain simulator for generalized extreme-value variates. This paper investigates the empirical consequences of nonlinear RUMs in the case of sportfishing modal choice, while refining and contrasting the available methods for welfare estimation. NONLINEAR INCOME EFFECTS IN RANDOM UTILITY MODELSJoseph A. Herriges and Catherine L. Kling* Abstract -Random utility models (RUMs) are used in the literature to model consumer choices from among a discrete set of alternatives, and they typically impose a constant marginal utility of income on individual preferences. This assumption is driven partially by the difficulty of constructing welfare estimates in models with nonlinear income effects. Recently, McFadden (1995) developed an algorithm for computing these welfare impacts using a Monte Carlo Markov chain simulator for generalized extreme-value variates. This paper investigates the empirical consequences of nonlinear RUMs in the case of sportfishing modal choice, while refining and contrasting the available methods for welfare estimation.

show abstract

“…A related area of concern is decisions over appropriate nesting structures in multiple site recreation demand models. Kling & Thomson [10] show that parameter estimates depend on both nesting structure and estimation method (sequential or Full Information Maximum Likelihood), whilst Herriges & Kling [11] report the sign and size of bias from inappropriate nesting structures and analyze the ability of conventional goodness-of-fit tests to identify the best model.…”

Section: Utility Specification Issues In the Environmental Valuation mentioning

confidence: 99%

How wrong can you be? Implications of incorrect utility function specification for welfare measurement in choice experiments

Figuerola¹,

Hanley

Riera³

2011

Journal of Environmental Economics and Management

View full text Add to dashboard Cite

Despite the vital role of the utility function in welfare measurement, the implications of working with incorrect utility specifications have been largely neglected in the choice experiments literature. This paper addresses the importance of specification with a special emphasis on the effects of mistaken assumptions about the marginal utility of income. Monte Carlo experiments were conducted using different functional forms of utility to generate simulated choices. Multi-Nomial Logit and Mixed Logit models were then estimated on these choices under correct and incorrect assumptions about the true, underlying utility function. Estimated willingness to pay measures from these choice modelling results are then compared with the equivalent measures directly calculated from the true utility specifications. Results show that for the parameter values and functional forms considered, a continuous-quadratic or a discrete-linear attribute specification is a good option regardless of the true effects the attribute has on utility.We also find that mistaken assumptions about preferences over costs magnify attribute mis-specification effects.

show abstract

The Implications of Model Specification for Welfare Estimation in Nested Logit Models

Cited by 80 publications

References 21 publications

Modeling Recreation Demand When the Access Point Is Unknown

Modeling Recreation Demand When the Access Point Is Unknown

Nonlinear Income Effects in Random Utility Models

How wrong can you be? Implications of incorrect utility function specification for welfare measurement in choice experiments

Contact Info

Product

Resources

About