Catálogo de publicaciones - libros

Dichotomous choice contingent valuation involves a binary yes/no question that can be followed by a subsequent question in order to obtain more information from the respondent, and leading to more efficient welfare estimates. Estimation methods for these data have mainly focused on classical maximum likelihood. In this paper we study possible improvements utilising a Bayesian MCMC approach to model this type of data. The classical and Bayesian approaches are compared with a Monte Carlo simulation experiment. The results show that the Bayesian approach improves the performance of the model, particularly with relatively small samples.

A Monte Carlo analysis is conducted to assess the validity of the bivariate modeling approach for detection and correction of different forms of elicitation effects in double bound contingent valuation data. Alternative univariate and bivariate models are applied to several simulated data sets, each one characterized by a specific elicitation effect, and their performance is assessed using standard selection criteria. The bivariate models include the standard bivariate probit model, and an alternative specification, based on the Copula approach to multivariate modeling, which is shown to be useful in cases where the hypothesis of normality of the joint distribution is not supported by the data. It is found that the bivariate approach can effectively correct elicitation effects while maintaining an adequate level of efficiency in the estimation of the parameters of interest.

Environmental economists have advocated the use of choice modelling in environmental valuation. Standard approaches employ choice sets including one alternative depicting the status-quo, yet the effects of explicitly accounting for systematic differences in preferences for non status-quo alternatives in the econometric models are not well understood. We explore three different ways of addressing such systematic differences using data from two choice modelling studies designed to value the provision of environmental goods. Preferences for change versus status-quo are explored with standard conditional logit with alternative-specific constant for status-quo, nested logit and a less usual mixed logit error component specification (kernel logit). Our empirical results are consistent with the hypothesis that alternatives offering changes from status-quo do not share the same preference structure as status-quo alternatives, as found by others in the marketing literature, in the environmental economic literature and in food preference studies. To further explore the empirical consequences of such mis-specification we report on a series of Monte Carlo experiments. Evidence from the experiments indicates that the expected bias in estimates ignoring the status-quo effect is substantial, and—more interestingly—that error component specifications with status-quo alternative specific-constant are efficient even when biased. These findings have significant implications for practitioners and their stance towards the strategies for the econometric analysis of choice modelling data for the purpose of valuation.

Although it is well established that dynamically optimal policies should be “closed loop” so that policies take into account changing conditions of a system, it is rare for such optimization to actually be carried out in large-scale simulation models. Computational limitations remain a major barrier to the study of dynamically optimal policies. Since the size of dynamic optimization problems grows approximately geometrically with the state space, this problem will continue to inhibit the identification of dynamically optimal policies for the foreseeable future. In this chapter, we explore in detail the problem of solving dynamic optimization problems for large-scale simulation models and consider methods to work around the computational barriers. We show that a reasonable approach is to solve a small-scale problem to identify an approximate value function that can then be embedded directly in the simulation model to find approximately optimal time-paths. We present and compare two ways to specify the small-scale problem: a traditional “meta-modelling” approach, and a new “direct approach” in which the simulation model is embedded directly in the dynamic optimization algorithm. The methods are employed in a model of the Gulf of Mexico’s red snapper fishery and used to identify the dynamically optimal total allowable catch for the recreational and commercial sectors of the fishery.

In the analysis of dynamic decision problems, the vast majority of the literature has focused on normative aspects: what should a resource manager do to maximize a particular objective function? Rarely have resource economists attempted to answer the positive question: what decision problem does a resource manager actually solve? There are several candidate explanations for this emphasis on normative modeling, but in this chapter we take the perspective that although structural estimation of discrete dynamic decision problems is not especially difficult, it is difficult enough that most analysts require some explanation of why they should bother with it at all. We develop a general model of a discrete dynamic decision problem, distill it to a tractable form, and present the estimation methodology. We then provide an empirical example and investigate the implications of using a reduced-form static model of behavior when the underlying data-generating process is dynamic.

The role of Monte Carlo methods in environmental economics and its subdisciplines, has increased in importance during the past several years. Due to the increasing power of computers and the development of sophisticated software, Monte Carlo and other computer-based simulation methods have emerged and established themselves as a third approach for advancing environmental and resource economics along side with traditional theory and empirical research. In this paper we review the contribution in environmental and resource economics of Monte Carlo method applications, illustrate guidelines for Monte Carlo results to be effectively and accurately communicated to and independently reproduced by other researchers, and survey the main methods and software options for executing Monte Carlo experiments.

In environmental economics, numerical simulation using random draws is the method most commonly used to estimate joint probabilities of individual choices in discrete-choice, random-parameters models. This paper compares simulation to another method of estimation, Gaussian quadrature, on the basis of speed and accuracy. The comparison is done using stated preference data consisting of the answers to choice questions for fishing in Green Bay, a large bay on Lake Michigan. Each sampled individual chose between a pair of Green Bay scenarios with different fishing conditions. Quadrature is found to be as accurate as simulation based on random draws, but Gaussian quadrature attains stability in estimated parameters considerably faster.

This study investigates issues surrounding the nature and importance of simulation noise when using maximum simulated likelihood methods in bivariate tobit estimation of panel data. The application presented considers land use decisions made by nomadic herders in northern Kenya. The study focuses on issues of parameter instability arising from the use of simulation methods to control for an unobserved household specific effect. It is found that parameters are more stable across estimation runs for variables for which there is a higher degree of within household variability and when the parameter is estimated with a higher degree of precision in the initial run. The study also finds that there is less variability in simulating estimation results when different draws are used to simulate results of a given estimation run than when results from different estimation runs generated by using different draws are used for simulation. It is also found that simulation noise does not have a large impact on a main policy finding of the estimation and simulation: reducing risk of accessing remote grazing areas can improve the spatial distribution of grazing pressure and thus address localized degradation and a failure to provide security can lead to environmental degradation.