On parallelizable Markov chain Monte Carlo algorithms with waste-recycling

Parallelizable Markov chain Monte Carlo (MCMC) generates multiple proposals and parallelizes the evaluations of the likelihood function on different cores at each MCMC iteration. Inspired by Calderhead (Proc Natl Acad Sci 111(49):17408–17413, 2014), we introduce a general ‘waste-recycling’ framework for parallelizable MCMC, under which we show that using weighted samples from waste-recycling is preferable to resampling in terms of both statistical and computational efficiencies. We also provide a simple-to-use criteria, the generalized effective sample size, for evaluating efficiencies of parallelizable MCMC algorithms, which applies to both the waste-recycling and the vanilla versions. A moment estimator of the generalized effective sample size is provided and shown to be reasonably accurate by simulations.     

Projecting age-specific fertility rates by using time-series methods

This paper discusses twostochastic time-series models developed recently for projecting age-specific fertility rates: the CARIMA model and the APC-ARIMA model. The forecasting performance of both models is examined using Dutch data. Alternatively, adeterministic time-series model is presented in which the age pattern of changes in the age-specific fertility rates between successive years is described by a cubic spline function. The model is capable of describing widely varying patterns. The model is applied to age-specific fertility rates for four countries: the Netherlands, England and Wales, Sweden and Australia.Cet article traite de deux modèles stochastiques qui utilisent des séries chronologiques: le modèle CARIMA et le modèle APC-ARIMA qui ont été développés récemment pour projeter les taux de fécondité par âge. La qualité des projections réalisées avec ces deux modèles est testée sur des données hollandaises. Simultanément, un modèle déterministe est présenté dans lequel les changements des taux de fécondité par âge entre années successives sont décrits par une fonction du troisième degré. Ce modèle est capable de décrire des situations qui varient dans un très large domaine. Il est appliqué aux taux de fécondité par âge de quatre pays: la Hollande, l'Angleterre et le Pays de Galles, la Suède et l'Australie.

---

The views expressed in this paper are those of the author and do not necessarily reflect the policies of the Netherlands Central Bureau of Statistics.     

The social power of algorithms

ABSTRACT

This article explores the questions associated with what might be thought of as the social power of algorithms. The article, which introduces a special issue on the same topic, begins by reflecting on how we might approach algorithms from a social scientific perspective. The article is then split into two sections. The first deals with the issues that might be associated with an analysis of the power of the algorithms themselves. This section outlines a series of issues associated with the functionality of the algorithms and how these functions are powerfully deployed within social world. The second section then focuses upon the notion of the algorithm. In this section, the article argues that we need to look beyond the algorithms themselves, as a technical and material presence, to explore how the notion or concept of the algorithm is also an important feature of their potential power. In this section, it is suggested that we look at the way that notions of the algorithm are evoked as a part of broader rationalities and ways of seeing the world. Exploring the notion of the algorithm may enable us to see how algorithms also play a part in social ordering processes, both in terms of how the algorithm is used to promote certain visions of calculative objectivity and also in relation to the wider governmentalities that this concept might be used to open up.     

Explaining differences in unemployment benefit takeup between labour migrants and Dutch native workers

The large increase in Eastern European migrants entering the Dutch labour market has led to concerns about their potential claim on Dutch unemployment benefits. We use a decomposition analysis to investigate differences in uptake of unemployment benefits between migrants and native Dutch employees by analysing register data for all employees in the Netherlands in 2015. The results show that Eastern European migrants, similar to other migrants, receive unemployment benefit more often than native Dutch employees. This difference can be largely ascribed to job characteristics. The inclusion of unemployment risk in the analysis reveals that non-working migrants are much less likely to receive unemployment benefits than Dutch natives.     

What works for whom? An analysis of active labour market programmes in Norway

This article examines whether some active labour market programmes (ALMP) targeted at social assistance recipients are more effective for some groups than for others in increasing self-sufficiency , i.e. gross earnings . The study focuses on the interaction between two types of programmes – employment programmes and training programmes – and the social charac-teristics of the participants. The data are derived from the pool of the entire population of social assistance recipients registered in Norway in 1995. The study adopts a quasi-experimental design. To handle selection bias, a matching procedure based on a propensity score approach is applied. Training programmes yield positive outcomes overall and for subgroups, whereas employment programmes produce no significant effects overall. However, for employment programmes there is an interaction effect between the likelihood of participation and outcome: beneficiaries with social characteristics (e.g. age, education, work experience) that are associated with a medium chance of participating experience a positive and significant increase in earnings. For those with lower and higher chances, the effect is negative. This points to the importance of conducting stratified analyses in effect evaluations. Thus, the results are also likely to be more relevant to policy makers.     

Bootstrapping a hedonic price index: experience from used cars data

Every hedonic price index is an estimate of an unknown economic parameter. It depends, in practice, on one or more random samples of prices and characteristics of a certain good. Bootstrap resampling methods provide a tool for quantifying sampling errors. Following some general reflections on hedonic elementary price indices, this paper proposes a case-based, a model-based, and a wild bootstrap approach for estimating confidence intervals for hedonic price indices. Empirical results are obtained for a data set on used cars in Switzerland. A simple and an enhanced adaptive semi-logarithmic model are fit to monthly samples, and bootstrap confidence intervals are estimated for Jevons-type hedonic elementary price indices.     

A new nonparametric estimator for the mean of the selected population

A modified bootstrap estimator of the mean of the population selected from two populations is proposed which is a convex combination of the two sample means, where the weights are random quantities. The estimator is shown to be strongly consistent. The small sample behavior of the estimator is investigated and compared with some competitors by means of Monte Carlo studies. It is found that the newly proposed estimator has smaller mean squared error for a wide range of parameter values.