Fast parallel $$\alpha $$-stable distribution function evaluation and parameter estimation using OpenCL in GPGPUs

\(\alpha \)-Stable distributions are a family of probability distributions found to be suitable to model many complex processes and phenomena in several research fields, such as medicine, physics, finance and networking, among others. However, the lack of closed expressions makes their evaluation analytically intractable, and alternative approaches are computationally expensive. Existing numerical programs are not fast enough for certain applications and do not make use of the parallel power of general purpose graphic processing units. In this paper, we develop novel parallel algorithms for the probability density function and cumulative distribution function—including a parallel Gauss–Kronrod quadrature—, quantile function, random number generator and maximum likelihood estimation of \(\alpha \)-stable distributions using OpenCL, achieving significant speedups and precision in all cases. Thanks to the use of OpenCL, we also evaluate the results of our library with different GPU architectures.     

The gender gap in pensions in Latin America

Lower female lifetime labour market participation rates, greater interruptions during their working lives, and wage gaps contribute to create gender gaps in pensions at the time of retirement. The design of social security systems may reinforce or attenuate these gaps. This article provides new evidence on gender gaps in access to pensions and in pension income in four Southern Cone countries in Latin America and analyses their evolution between 2000 and 2013, showing significant improvements in both gaps, with differential patterns by countries. The decrease in the gender gap in pension income has been particularly significant in Argentina and Brazil. In both cases, the largest increases in pension values during the period correspond to the lowest income percentiles, where women are overrepresented. The application of redistributive policies in these countries, aimed at reducing poverty and inequality but not necessarily focused on gender equity, has had positive and probably unintended consequences in terms of reduction in gender gaps in pensions.     

Bayesian robustness of the compound Poisson distribution under bidimensional prior: an application to the collective risk model

The distribution of the aggregate claims in one year plays an important role in Actuarial Statistics for computing, for example, insurance premiums when both the number and size of the claims must be implemented into the model. When the number of claims follows a Poisson distribution the aggregated distribution is called the compound Poisson distribution. In this article we assume that the claim size follows an exponential distribution and later we make an extensive study of this model by assuming a bidimensional prior distribution for the parameters of the Poisson and exponential distribution with marginal gamma. This study carries us to obtain expressions for net premiums, marginal and posterior distributions in terms of some well-known special functions used in statistics. Later, a Bayesian robustness study of this model is made. Bayesian robustness on bidimensional models was deeply treated in the 1990s, producing numerous results, but few applications dealing with this problem can be found in the literature.     

The t-information and its use in multivariate problems and time-series analysis

The second moment of recently introduced scalar inference function can be viewed as generalized Fisher information of the continuous probability distributions. In this paper we call it the t-information and give some possible applications of the new concept.     

Link-tracing sampling with an initial sequential sample of sites: Estimating the size of a hidden human population

We proposed a modification to the variant of link-tracing sampling suggested by Félix-Medina and Thompson [M.H. Félix-Medina, S.K. Thompson, Combining cluster sampling and link-tracing sampling to estimate the size of hidden populations, Journal of Official Statistics 20 (2004) 19–38] that allows the researcher to have certain control of the final sample size, precision of the estimates or other characteristics of the sample that the researcher is interested in controlling. We achieve this goal by selecting an initial sequential sample of sites instead of an initial simple random sample of sites as those authors suggested. We estimate the population size by means of the maximum likelihood estimators suggested by the above-mentioned authors or by the Bayesian estimators proposed by Félix-Medina and Monjardin [M.H. Félix-Medina, P.E. Monjardin, Combining link-tracing sampling and cluster sampling to estimate the size of hidden populations: A Bayesian-assisted approach, Survey Methodology 32 (2006) 187–195]. Variances are estimated by means of jackknife and bootstrap estimators as well as by the delta estimators proposed in the two above-mentioned papers. Interval estimates of the population size are obtained by means of Wald and bootstrap confidence intervals. The results of an exploratory simulation study indicate good performance of the proposed sampling strategy.     

Residential Remittances and Food Security in the Upper West Region of Ghana

In recent years, out migration from the Upper West Region to the southern belt of Ghana for farming has become commonplace. The natural question that has arisen is: what is the potential impact of remittances from this migration pattern on food security in the region? Using multivariate ordered logistic regression this study assesses the linkage between remittances and household food security (derived using the HFIAS) among urban and rural households (n=1,438) in the region. The findings show that urban remittance‐receiving households and rural remittance and non‐remittance receiving households were more likely (OR=2.44, p<0.05; OR=2.46, p<0.001; and OR=1.49, p<0.1, respectively) to report being more severely food‐insecure than urban non‐remittance receiving households. The findings demonstrate that household strategies such as migration and remittances on their own are not sufficient to ameliorate the precarious food insecurity situation of the region. The study calls for development of alternative livelihoods in the region.     

Stagnation only on the surface? The implications of skill and family responsibilities for the gender wage gap in Sweden, 1974–2010

The wage differential between women and men persists in advanced economies despite the inflow of women into qualified occupations in recent years. Using five waves of the Swedish Level‐of‐Living Survey (LNU), this paper explores the gender wage gap in Sweden during the 1974–2010 period overall and by skill level. The empirical analyses showed that the general gender wage gap has been nearly unchanged for the past 30 years. However, the gender difference in wage in less qualified occupations fell considerably, whereas the gender pay gap remained stable for men and women in qualified occupations. The larger significance of family responsibilities for wages in qualified occupations is one likely explanation for this result.     

PEER PRESSURE: EXPERIMENTAL EVIDENCE FROM RESTROOM BEHAVIOR

We provide experimental evidence on the effect of peer pressure on individual behavior. Specifically, we study the effect of being exposed to an observer in a public restroom on handwashing and urinal flushing behavior. Our estimates show that being exposed to an observer increases the probability of handwashing by 13 percentage points and the probability of urinal flushing by 15 percentage points. Given that handwashing and urinal flushing have social benefits that exceed individual benefits, our findings provide support for peer pressure as an additional way of addressing the social suboptimality arising from externalities. (JEL C91, C93)     

Competing Priorities: Partner-Specific Relationship Characteristics and Motives for Condom Use Among At-Risk Young Adults

Condoms protect against human immunodeficiency virus (HIV)/sexually transmitted infections (STIs) and unintended pregnancy and are essential to sexual health efforts targeting young adults, who are disproportionately affected by both outcomes. Understanding condom use motives is critical to increasing condom use. Research in this area is limited, particularly regarding the roles of partners and relationship factors. Using a longitudinal sample of 441 young adults and 684 reported partnerships we examined associations between relationship factors and condom use motives (pregnancy prevention, disease prevention, or dual protection). Simultaneous multilevel models identified variables associated with motives; level-specific models identified the levels (individual, partnership, time) variables impacted motives. Participants reported choosing condoms for pregnancy prevention, disease prevention, and dual protection in 51%, 17%, and 33% of partnerships, respectively. Partner-specific factors varied, to a differing degree, across the three levels. Seven variables (duration, condom self-efficacy, commitment, sexual decision-making, power, and vulnerability to harm [HIV/STIs] and pregnancy) distinguished condom use motives. The level of this association varied but was most pronounced at the partner and individual levels. Researchers and practitioners should consider the impact of both individual- and partner-level factors on condom use motives, in both research and sexual health programs.