Selecting random parameters in discrete choice experiment for environmental valuation: A simulation experiment

Abstract

Abstract.This paper examines the various tests commonly used to select random parameters in choice modelling. The most common procedures for selecting random parameters are: the Lagrange Multiplier test as proposed by McFadden and Train (2000), the t-statistic of the deviation of the random parameter and the log-likelihood ratio test. The identification of random parameters in other words the recognition of preference heterogeneity among population is based on the fact that an individual makes a choice depending on her/his: tastes, perceptions and experiences. A simulation experiment was carried out based on a real choice experiment and the results indicated that the power of these three tests depends importantly on the spread and type of the tested parameter distribution.

Motivations and attitudes influence farmers' willingness to participate in biodiversity conservation contracts

“…RPL models thus take into account heterogeneity of the parameter values among respondents. Compared to multinomial logit models, they are therefore behaviourally more appropriate and more policy relevant underpinned (Jaeck and Lifran, 2014;Mariel et al, 2013;Marsh, 2012;Schulz et al, 2014). Simulations are required in RPL models to approximate parameters as preference coefficients are not directly observable.…”

The Cost of Mediterranean Sea Warming and Acidification: A Choice Experiment Among Scuba Divers at Medes Islands, Spain

“…The latter term is identically and independently distributed (IID) as extreme value. The parameter β is homogeneous across individuals and is considered as a weight measure explaining the contribution of each random variable to the overall utility of an alternative (Mariel et al 2013).…”

“…This allows capturing heterogeneity of preferences as a function of individuals' characteristics (Train 1998;Mariel et al 2013). In the RPL specification represented in Eq.…”

“…In the RPL specification represented in Eq. (2), the coefficient vector is defined as (β + η i j ), where β represents the mean attribute utility weights in the population, while η stands for the individual deviation from the mean (Train 1998;Mariel et al 2013).…”

Economic Valuation of Health Services in the Basque Country

“…La gran mayoría de los autores no aplica este test para contrastar la posible aleatoriedad de los parámetros y utiliza en su lugar los estadísticos t de las desviaciones típicas de los parámetros aleatorios. Mariel et al (2013), sin embargo, demuestran que la aplicación de este procedimiento debe evitarse porque presenta el error tipo I muy elevado y la aplicación del procedimiento propuesto por McFadden y Train (2000) se hace, por tanto, necesaria.…”