A Comparison of Bayesian and Sampling Theory Inferences in a Probit Model

HE PROBIT MODEL IS A POPULAR DEVICE for explaining binary choice decisions in econometrics. It has been used to describe choices such as labor force participation, travel mode, home ownership, and type of education. These and many more examples can be found in papers by Amemiya (1981) and Maddala (1983). Given the contribution of economics towards explaining such choices, and given the nature of data that are collected, prior information on the relationship between a choice probability and several explanatory variables frequently exists. Bayesian inference is a convenient vehicle for including such prior information. Given the increasing popularity of Bayesian inference it is useful to ask whether inferences from a probit model are sensitive to a choice between Bayesian and sampling theory techniques. Of interest is the sensitivity of inference on coefficients, probabilities, and elasticities. We consider these issues in a model designed to explain choice between fixed and variable interest rate mortgages. Two Bayesian priors are employed: a uniform prior on the coefficients, designed to be noninformative for the coefficients, and an inequality restricted prior on the signs of the coefficients. We often know, a priori, whether increasing the value of a particular explanatory variable will have a positive or negative effect on a choice probability. This knowledge can be captured by using a prior probability density function (pdf) that is truncated to be positive or negative. Thus, three sets of results are compared:those from maximum likelihood (ML) estimation, those from Bayesian estimation with an unrestricted uniform prior on the coefficients, and those from Bayesian estimation with a uniform prior truncated to accommodate inequality restrictions on the coefficients.

Migration of Husbands, Remittances and Agricultural Production: Impacts when Wives Head Households in Rural Kenya

This paper uses data collected from migrants' wives in the Nyeri district of Kenya. The main objective is to determine whether migration and remittances contribute to the development of agriculture. Our results suggest that most migrants are pushed out of rural areas, belong to the group of low-paid workers in urban areas, send little and irregular remittances to their wives back in rural areas and that these remittances are mainly used for consumption purposes and do not contribute to any significant development in agriculture. Our results also indicate that altruism or social obligation might be the main reason for migrants sending remittances back to their rural wives.

Children and economic development: Family size, gender preferences and human capital formation - theory and Indian cases

In the light of Gary Becker's economic theory of the family, considers how economic cost and benefit factors can influence the size of families that parents decide to have. Some support for the importance of such factors is found from results of structured interviews with wives in Kondh-dominated villages in western Orissa. These results are at variance with the hypothesis of Malthus about population growth. Factors that may alter the optimal family size as development proceeds are discussed. It is found in our sampling that, on the whole, there is a preference for daughters rather than sons although this is not as strong in the Kondh-dominated villages as in poor villages in the Santal tribal belt of West Bengal. While in the Kondh-dominated villages some discrimination in access to education in favour of boys compared to girls is present, little such or no such discrimination occurs in relation to access to food and medical attention. In the villages surveyed in the West Bengal Santal tribal belt, discrimination in favour of boys is more pronounced than in the Kondh-dominated area in Orissa. While economic considerations help to explain gender discrimination between boys and girls, we find that social and cultural factors also play a major role. Parents in a similar economic situation seem to display substantially different patterns of gender discrimination between children depending on their social and cultural content. It seems that the extent to which economic theories of the family explain family preferences and behaviour depend significantly on the social and cultural context in which they are to be applied.

Notes on market failure and the Paretian (Kaldor-Hicks) relevance and irrelevance of unfavourable externalities

The following are notes that have been distributed by me over the last few years to students in Environmental Economics at The University of Queensland. They give particular attention to whether externalities are Pareto or Kaldor-Hicks relevant from a policy point of view. Externalities are Kaldor-Hicks or Pareto irrelevant if no change is possible for which gainers could compensate losers. Both absolute and marginal externalities may be Kaldor-Hicks relevant. Infra-marginal negative externalities are often, but not always, Kaldor-Hicks irrelevant. There are at least two cases where such externalities can be relevant. First, the absolute impact of the negative externality may be so great that the source of the externality should be eliminated. Secondly, if the externality arises from production, its nature may depend on the type of production technique adopted. Although for the technique adopted, an infra-marginal negative externality occurs. That is Paretian irrelevant given that choice of this technique is the only available possibility, alternative techniques may actually be available in practice. Some of these may generate even smaller total external effects and be socially preferable. Both cases are outlined and illustrated in these notes. The analysis reveals the dangers of relying on marginalism for deciding on environmental policy. Total (external) effects are often of great social and economic importance and appropriate social choices cannot be made on the basis of marginalism alone.