A fast splitting procedure for classification trees

This paper provides a faster method to find the best split at each node when using the CART methodology. The predictability index is proposed as a splitting rule for growing the same classification tree as CART does when using the Gini index of heterogeneity as an impurity measure. A theorem is introduced to show a new property of the index : the for a given predictor has a value not lower than the for any split generated by the predictor. This property is used to make a substantial saving in the time required to generate a classification tree. Three simulation studies are presented in order to show the computational gain in terms of both the number of splits analysed at each node and the CPU time. The proposed splitting algorithm can prove computational efficiency in real data sets as shown in an example.     

A log-logistic regression model for a transition rate with a starting threshold

The paper presents a new type of parametric transition rate model that is particularly suitable for studying household and union formation. In such studies, it is often not easy to determine the moment at which individuals actually enter the population at risk - or at least when the risk begins to become socially important. Even in the presence of regulations stipulated by law, one might be interested in studying ‘social’ timetables. We assume that there is a constant rate for any duration and that after a certain threshold point a log-logistic rate is added, with this threshold as its time origin. This can be justified in a behavioural sense by assuming that random and social transitions arise from separate processes. We then apply the model to union formation in Italy and show how threshold and intensity effects generated by theoretical hypotheses can be revealed.     

DYNASTIC MANAGEMENT

The most striking difference in corporate‐governance arrangements between rich and poor countries is that the latter rely much more heavily on the dynastic family firm, where ownership and control are passed on from one generation to the other. We argue that if the heir to the family firm has no talent for managerial decision making, dynastic management is a failure of meritocracy that reduces a firm's total factor productivity (TFP). We present a simple model that studies the macroeconomic causes and consequences of dynastic management. In our model, the incidence of dynastic management depends, among other factors, on the imperfections of contractual enforcement. A plausible calibration suggests that, via dynastic management, poor contract enforcement may be a substantial contributor to observed cross‐country differences in aggregate TFP. (JEL O43, O47, G32)     

Beyond deportation: researching the control of outward mobility using a space of flows logic

Manuel Castells (1996) famously argued that human processes are increasingly operating according to the logic of flows and it has now become commonplace to analyse movements of people, information and commodities in terms of flows. However, scholars have been slow to capture the dynamics of border enforcement practices in these terms. In this article, we argue that ‘deportation’ can best be understood, not as a discrete practice that is unidirectional, territorial and wholly controlled by individual states, but as a range of diverse practices used by states (and sometimes undermined by other parties) to try to control the circulation of people within a dynamic supra‐national space. By focusing on ‘mobility control continuums’ operating in selected countries at the peripheries of Europe, we capture the dynamics of state intervention in trans‐border flows and thereby contribute towards developing concepts and methodologies for the criminological study of border controls that are ‘sensitive to the complexities of the global’ (Aas 2007).     

Focused Information Criterion for Capture–Recapture Models for Closed Populations

Abstract. We propose a criterion for selecting a capture–recapture model for closed populations, which follows the basic idea of the focused information criterion (FIC) of Claeskens and Hjort. The proposed criterion aims at selecting the model which, among the available models, leads to the smallest mean‐squared error (MSE) of the resulting estimator of the population size and is based on an index which, up to a constant term, is equal to the asymptotic MSE of the estimator. Two alternative approaches to estimate this FIC index are proposed. We also deal with multimodel inference; in this case, the population size is estimated by using a weighted average of the estimates coming from different models, with weights chosen so as to minimize the MSE of the resulting estimator. The proposed model selection approach is compared with more common approaches through a series of simulations. It is also illustrated by an application based on a dataset coming from a live‐trapping experiment.