Estimation of contingency tables in complex survey sampling using probabilistic expert systems

In this paper we explore the possibility to use a particular class of models, known as probabilistic expert systems, to define two classes of estimators of a contingency table in case of stratified sampling designs. The two classes are characterized by the different role of the sampling design: in the first, the sampling design is treated as an additional variable; in the second, it is used only for estimation purposes by means of the survey weights. The bias/variance trade off of these estimators is analyzed and the consequences of model misspecification are illustrated. Furthermore, it is shown that the Horvitz–Thompson estimator belongs to both classes of estimators. It comes out that the Horvitz–Thompson estimator is almost always inefficient but robust. Monte Carlo simulations illustrate the efficiency of the proposed estimators.     

Analysis and correction of bias in Total Decrease in Node Impurity measures for tree-based algorithms

Variable selection is one of the main problems faced by data mining and machine learning techniques. These techniques are often, more or less explicitly, based on some measure of variable importance. This paper considers Total Decrease in Node Impurity (TDNI) measures, a popular class of variable importance measures defined in the field of decision trees and tree-based ensemble methods, like Random Forests and Gradient Boosting Machines. In spite of their wide use, some measures of this class are known to be biased and some correction strategies have been proposed. The aim of this paper is twofold. Firstly, to investigate the source and the characteristics of bias in TDNI measures using the notions of informative and uninformative splits. Secondly, a bias-correction algorithm, recently proposed for the Gini measure in the context of classification, is extended to the entire class of TDNI measures and its performance is investigated in the regression framework using simulated and real data.     

A parametric model fitting time to first event for overdispersed data: application to time to relapse in multiple sclerosis

In this article, we propose a parametric model for the distribution of time to first event when events are overdispersed and can be properly fitted by a Negative Binomial distribution. This is a very common situation in medical statistics, when the occurrence of events is summarized as a count for each patient and the simple Poisson model is not adequate to account for overdispersion of data. In this situation, studying the time of occurrence of the first event can be of interest. From the Negative Binomial distribution of counts, we derive a new parametric model for time to first event and apply it to fit the distribution of time to first relapse in multiple sclerosis (MS). We develop the regression model with methods for covariate estimation. We show that, as the Negative Binomial model properly fits relapse counts data, this new model matches quite perfectly the distribution of time to first relapse, as tested in two large datasets of MS patients. Finally we compare its performance, when fitting time to first relapse in MS, with other models widely used in survival analysis (the semiparametric Cox model and the parametric exponential, Weibull, log-logistic and log-normal models).     

Changes in union status during the transition to parenthood in eleven European countries, 1970s to early 2000s

Couples who have children are increasingly likely to have lived together without being married at some point in their relationship. Some couples begin their unions with cohabitation and marry before first conception, some marry during pregnancy or directly after the first birth, while others remain unmarried 3 years after the first birth. Using union and fertility histories since the 1970s for eleven countries, we examine whether women who have children in unions marry, and if so, at what stage in family formation. We also examine whether women who conceive when cohabiting are more likely to marry or separate. We find that patterns of union formation and childbearing develop along different trajectories across countries. In all countries, however, less than 40 per cent of women remained in cohabitation up to 3 years after the first birth, suggesting that marriage remains the predominant institution for raising children.     

Here or There? Shifting Meanings and Practices in Mother–Child Relationships across Time and Space

In this paper, I analyse the changes that mothers and children experience in their relationship due to the physical separations and reunions entailed by the international migration process. I argue that the different geographical configurations that migrant families take over time are the outcome of a negotiation of care responsibility and desired geographies of family life, and are accompanied by changing meanings and practices in intimate relationships: the location of care relationships is influenced by the relatives' capacity both to take part in family negotiations as well as to overcome the constraints imposed by policies. Time is relevant because it leads to shifting meanings and practices of transnational family life, as well as to the changing role of children in the family.     

Kuznets Prize for Makoto Hirazawa and Akira Yakita

News

Kuznets Prize for Makoto Hirazawa and Akira Yakita     

DOES DIVORCE LAW MATTER?

In this paper we derive an explicit model of negotiations between spouses when unconstrained transfers are possible only in case of separation. We show that inefficient separation may occur in equilibrium even under consensual divorce law. This provides theoretical support for the view that changes in social norms rather than in legislation may be responsible for increasing divorce rates. (JEL: C78, J120)     

Technical Progress,Life of Capital and Employment

This paper examines the hypothesis of a quickening of capital replacement, triggered by the acceleration of technical progress. We present a microeconomic model of replacement processes inspired by the “unbalanced growth” school, highlighting the role of productivity in determining the growth path. The model distinguishes between two types of enterprise, stagnant and progressive, on the basis of productivity trends and shows that prolonged acceleration in the rate of technical progress may be sufficient reason for a rational entrepreneur to make replacements earlier, thus approaching a system of continuous restructuring. We then discuss the main theoretical implications for aggregate employment and the fundamental questions of measurability of the effects. To this end, we undertake an empirical test of aggregate data for the manufacturing sector in four leading economies (the United States, France, Germany, the United Kingdom) from 1970 to 1991. We estimate labour demand functions in which the traditional explanatory variables are supplemented by proxies that capture the effect of a shortening of the life of capital equipment (the replacement rate and the ratio between depreciation and gross fixed capital formation). The empirical evidence confirms the shortening of the average lifetime of capital and is consistent with the main implications of the microeconomic model. The results show, at least indirectly, a negative correlation between an increase in speed of technical progress and the growth of employment.     

The Role of Family Ties in the Labour Market. An Interpretation Based on Efficiency Wage Theory

By casual empiricism, it seems that many firms take explicit account of the family ties connecting workers, often hiring individuals belonging to the same family or passing jobs on from parents to their children. This paper makes an attempt to explain this behaviour by introducing the assumption of altruism within the family and supposing that agents maximize a family utility function rather than an individual one. This hypothesis has been almost ignored in the analysis of the relationship between employers and employees. The implications of this assumption in the efficiency wage models are explored: by employing members of the same family, firms can use a (credible) harsher threat — involving a sanction for all the family’s members in case of one member’s shirking — that allows them to pay a lower efficiency wage. On the other hand, workers who accept this agreement exchange a reduction in wage with an increase in their probability of being employed: this can be optimal in a situation of high unemployment. Moreover, the link between parents and children allows the firm to follow a strategy that solves the problem of an individual’s finite time horizon by its making use of the family’s reputation.