My PhD dissertation was on “Bayesian Inference in Modeling Recreation Demand”. It focused on consistently estimating and evaluating the demand for and welfare derived from recreational sites. Two of the papers are published in the Environmental & Resource Economics and the American Journal of Agricultural Economics. I have written a couple of papers after that using similar methodology.
I have interest in methods that can help improve elicitation patterns of choice experiments and stated preference methods. My recent work on analyzing the cost of power outage and preference for electricity reliability incorporates uncertainty in preference elicitation formats using the elicited probability model by Manski. The ability to recover preference for electricity reliability is grounded in the principle of revealed preference and utility maximization. McFadden based the Random Utility maximization (RUM) model on relaxing the classical assumption of humus economicus by assuming that the unknown in the utility is only from the standpoint of the observer through unmeasured psychological factors. That is, there is no uncertainty in the utility from each alternative from the point of view of the respondent but the observer of the agent does not measure this utility appropriately because of measurement error, unobservables and specification errors. This assumption is the workhorse of the RUM model and the analysis of many discrete choice models. However, in many applications, there is reason to believe that the humus economicus assumption by itself may be flawed. For instance, when presented with a choice between outage during peak hours, off-peak hours or intermediate hours, some individuals may not be able to know which of these options will maximize their utility if they are not certain of the time when they will be out of the house especially for self employed individuals. These individuals may be willing to switch between different options at the time of making the decision but may not be able to convey these uncertainty when responding to a questionnaire.
Below is a summary of some of my other papers in this area:
Random Utility Maximization (RUM) models of recreation demand are typically plagued by limited information on environmental and other attributes characterizing the available sites in the choice set. To the extent that these unobserved site attributes are correlated with the observed characteristics and/or the key travel cost variable, the resulting parameter estimates and subsequent welfare calculations are likely to be biased. In this paper we develop a Bayesian approach to estimating a RUM model that incorporates a full set of alternative specific constants, insulating the key travel cost parameter from the influence of the unobserved site attributes. In contrast to estimation procedures recently outlined in Murdock (2006), the posterior simulator we propose (combining data augmentation and Gibbs sampling techniques) can be used in the more general mixed logit framework in which some parameters of the conditional utility function are random. Following a series of generated data experiments to illustrate the performance of the simulator, we apply the estimation procedures to data from the Iowa Lakes Project. In contrast to an earlier study using the same data (Egan et al. ), we find that, with the addition of a full set of alternative specific constants, water quality attributes no longer appear to influence the choice of where to recreate.
A Bayesian variable selection procedure is used to control for uncertainty in the specification of a recreational demand model. Specifically, we propose a model that draws on the Bayesian paradigm to integrate the variable selection process into model estimation and to reflect the accompanying uncertainty about which is the best specification in subsequent counterfactual predictions. The advantage of this procedure over previous non-Bayesian approaches is that it overcomes the problem of pre-testing in specification searches. In our application, evaluating demand for recreational lake usage in Iowa, we find clear evidence that site attributes, such as lakes size, handicap facilities and wake restrictions, do impact lake usage. There is also evidence that water quality matters in household recreation choices. Indeed, contrary to Abidoye et al. (Am J Agricult Econ, 2012 ), in which only a single functional form is considered, we find clear evidence that water quality matters, with posterior probability of less that 10 % associated with a model without any water quality variables. This suggests that the flexibility that the Bayesian variable selection model affords in capturing the linkage between recreation demand and site characteristics can be important.
A sample of 1,114 households in three countries located in the Limpopo River Basin was surveyed to investigate willingness to use (WTU) and willingness to pay (WTP) for different quality attributes of recycled water. The results indicate that 36% of the households are willing to use recycled water for potable uses and only 70% are willing to use it for non-potable use such as gardening. WTU recycled water was found to be associated with gender, age, education, occupation and perception of quality of water from the main source. Households are willing to pay 76% of the price of standard water for improved quality in recycled water with no colour.
A sample of U.S. consumers were surveyed in a choice based experiment in the Fall of 2005 and Spring 2006 to elicit consumers’ preferences for quality attributes in beef products. Based on the resulting data, a random coefficients logit model is estimated, and consumers’ willingness to pay for these quality attributes in beef products is obtained. The results indicate that consumers have strong valuation for traceability, grass-fed, and U.S. origin attributes in a standard rib-eye steak and are willing to pay a premium for these attributes.