Non linear parametric mode regression

In this article, we propose a semi-parametric mode regression for a non linear model. We use an expectation-maximization algorithm in order to estimate the regression coefficients of modal non linear regression. We also establish asymptotic properties for the proposed estimator under assumptions of the error density. We investigate the performance through a simulation study.     

An optimal k of kth MA-ARIMA models under a class of ARIMA model

In this article, we discuss finding the optimal k of (i) kth simple moving average, (ii) kth weighted moving average, and (iii) kth exponential weighted moving average based on simulated ARIMA(p, d, q) model. We run a simulation using the three above examining methods under specific conditions. The main finding is that 5th exponential weighted moving average (5th EWMA) ARIMA model is the best forecasting model among others, which means the optimal k = 5. For Turkish Telecommunications (TTKOM) stock market, real data reveal the similar results of simulation study.     

Gender differences in gambling preferences and problem gambling: a network-level analysis

Most gambling studies have a gender-blind research approach, although a large body of scientific evidence suggests that gambling in females is on the rise and that males and females have different gambling behaviours and experience specific gambling-related harm. This study addressed these gender differences using a network analysis, an innovative approach considering disorders/concepts as dynamic systems of interacting symptoms/items. Data on gambling activities, problem gambling, substance use and mental health were collected in a representative sample of French adult gamblers (n = 8805). The study capitalized on the network analysis directly to compare associations of specific gambling activities with gambling disorder symptoms separately for both genders. The network analysis revealed that problem gambling was strongly associated with gambling machines among females, whereas it was related to sports betting, poker and casino games among males. The networks that included substance use and mental health showed that substance use was related to specific gambling activities. These findings confirm the links between various gender-specific gambling patterns and problem gambling and suggest a need to consider these gender differences to improve prevention efforts. More broadly, the present study further supports the importance of gender differences for gambling research and policy.     

How leader humility helps teams to be humbler,psychologically stronger,and more effective: A moderated mediation model

We hypothesize that (a) the level of humility expressed by leaders predicts team performance through, serially, team humility and team PsyCap, and (b) the strength (i.e., consensus within the team) of the leader humility, team humility and team PsyCap moderates the paths of that hypothesized model. A sample comprising 82 teams (82 leaders; 332 team members) was collected. Team members reported leader humility, team humility and team PsyCap. Leaders reported team performance. To handle the risks of common method bias, each mediating path of the hypothesized model is based on data from two different subsamples within each team. Our model's most novel theoretical contribution is the (moderated mediated) connection between leader humility, collective humility, and team PsyCap, and this was consistently supported in our data. Our inconsistent findings dealing with the relationship between team PsyCap and performance is well established in the literature and our results in both sub-samples were in the theorized direction. The study contributes to understand why, how and when humble leaders are more effective.     

Les effets pervers du partage de la sécurité. Polices publiques et privées dans une gare et un centre commercial

This article focuses on controlling delinquency in a train station and a shopping center in a big French city. The increasing implication of private organizations in implementing security policies in public areas has spawned a literature about the “dangers of privatization”. As this fieldwork shows, private security officers and NGOs assume duties (related, for example to public relations or social policy) left vacant by police officers. Owing to this sharing of security, the police concentrate on its purely bureaucratic assignments and this has a negative impact on its legitimacy. A historical approach helps us understand how relations between the police and other parties have soured.     

Appreciating Impact: Evaluating Small Voluntary Organizations in the United Kingdom

Within the mixed economy of care in the United Kingdom there are debates about the ways in which impact can be evaluated, in order to shape funding and policy decisions. One of the tensions evident in this debate is whether the evaluation approach should reflect the perspectives and goals of the voluntary organizations and their members, or whether evaluation should reflect the wider goals of the whole system of provision. The former runs the risk of being insular and self-congratulatory, while the latter may be inappropriate and dismissive of achievements. This paper explores this tension by reporting on a study that used Appreciative Inquiry to evaluate 10 small-scale not-for-profit schemes for older people. The data indicated some unexpected and long-term impacts that demonstrated the distinctiveness of the sector. Subsequently the findings were mapped on to the “impact grid” developed by Wilding and Lacey (2003). While this was straightforward at the levels of individuals and interorganizationally, it was more difficult at the sector/community level, suggesting that more work needs to be done to bring these two perspectives together.     

Hierarchical Gaussian process mixtures for regression

As a result of their good performance in practice and their desirable analytical properties, Gaussian process regression models are becoming increasingly of interest in statistics, engineering and other fields. However, two major problems arise when the model is applied to a large data-set with repeated measurements. One stems from the systematic heterogeneity among the different replications, and the other is the requirement to invert a covariance matrix which is involved in the implementation of the model. The dimension of this matrix equals the sample size of the training data-set. In this paper, a Gaussian process mixture model for regression is proposed for dealing with the above two problems, and a hybrid Markov chain Monte Carlo (MCMC) algorithm is used for its implementation. Application to a real data-set is reported.     

A Class of Parametric Dynamic Survival Models

A class of parametric dynamic survival models are explored in which only limited parametric assumptions are made, whilst avoiding the assumption of proportional hazards. Both the log-baseline hazard and covariate effects are modelled by piecewise constant and correlated processes. The method of estimation is to use Markov chain Monte Carlo simulations Gibbs sampling with a Metropolis–Hastings step. In addition to standard right censored data sets, extensions to accommodate interval censoring and random effects are included. The model is applied to two well known and illustrative data sets, and the dynamic variability of covariate effects investigated.     

Union Stability and Stepfamily Fertility in Austria,Finland, France &; West Germany

In this paper, we test the hypothesis that unique effects of stepfamily composition on union fertility are confounded with differences between stepfamilies and couples without stepchildren in the risk of union disruption. We use birth and union histories from Fertility and Family Surveys in Austria, Finland, France and West Germany. The risks of a union birth and separation are modeled simultaneously, allowing for the potential effects of unobserved predispositions to have a child or to separate on the other event. We test hypotheses drawn from the value of first and second shared births to couples: Net of the couple’s combined parity, the birth risk will be greater if the child is (a) the first in a union, (b) the first biological child for one of the partners, or (c) the second child in a union.Henz, U. et Thomson, E., 2005. Stabilité des unions et fécondité des familles recomposées en Autriche, Finlande, France et Allemagne de l’ouest, Revue Européenne de Démographie, 21: 3–29.     

Capabilities and Well-Being: Evidence Based on the Sen–Nussbaum Approach to Welfare

One of the most significant theoretical contributions to welfare analysis across a range of disciplines has been the development of the capabilities framework by Sen and others. Motivated by the claim that freedom should play a key role in social evaluation, the capabilities framework suggests that we consider what it is that people are free to do, as well as what they actually do. Using data from the British Household Panel Survey in conjunction with a list of substantial values posited by Martha Nussbaum, we contribute to the operationalisation and testing of this approach. Specifically, we suggest that commonly used secondary data sources do provide some information about the capabilities people have and that this can be incorporated into models of (subjective) well-being such as those used by a growing number of labour and health economists. We find evidence that a wide range of capabilities exhibit statistically significant relations to well-being that the relations are complex and slightly different for men and women, and conclude with suggestions for future developments.