This code provides Bayesian estimation of the mixed logit model which accommodates bounded and correlated partworths (where "partworths" are coefficients of the attributes in the utility function). The model retains the numerical advantages of Bayesian procedures with correlated normals while also allowing for different bounded distributions of partworths (log-normal, normal censored from below at 0, Johnson's S_B distribution). The code allows variables with either fixed and/or random coefficients. The results are presented in the standard format for classically estimated models, namely by reporting the parameter estimates (posterior means) and their standard errors (posterior standard deviations).
White Halbert and Giacomini Raffaella