El estudio de las preferencias es uno de los temas menos tratados por la econometría aplicada. En el presente artículo se pretende realizar una aproximación translogarítimica de las preferencias por medio de la construcción de un Índice de Satisfacción de los Hogares (ISH) con base al...

In a discrete choice situation, information about the tastes of each sampled customer is inferred from estimates of the distribution of tastes in the population. First, maximum likelihood procedures are used to estimate the distribution of tastes in the population using the pooled data for all...

The choice of a specific distribution for random parameters of a discrete choice model is a critical issue in transportation analysis. Indeed, various pieces of research have demonstrated that an inappropriate choice of the distribution may lead to serious biases in model forecast and in the...

This paper provides necessary conditions for testing the local consistency of three-level nested logit models with random utility maximization. We find that for a model with two sub-nests per nest the conditions can lead to a substantial increase in the range of acceptable dissimilarity...

This paper develops a partial equilibrium model to account for stylized facts about the behavior of oligarchs, politically and economically strong conglomerates in transition and developing countries. The model predicts that oligarchs are more likely than other owners to invest in productivity...

We investigate the causes of civil war, using a new data set of wars during 1960-99. We test a `greed’ theory focusing on the ability to finance rebellion, against a`grievance’ theory focusing on ethnic and religious divisions, political repression and inequality. We find that greed...

This paper shows how to cope with a problem of model selection and simplification using the principle of coherence (Gabriel (1969): A procedure involving testing a set of models ought not accept a model while rejecting a more general model). The mathematical lattice theory is used to define a...

The simulation variance in the estimation of mixed logit parameters is found, in our application, to be lower with 100 Halton draws than with 1000 random draws. This finding confirms Bhat's (1999a) results and implies significant reduction in run times for mixed logit estimation. Further...

A relatively simple and convenient score test of normality in the bivariate probit model is derived. Monte Carlo simulations show that the small sample performance of the bootstrapped test is quite good. The test may be readily extended to testing normality in related models.