Model selection in quantile regression models

Lasso methods are regularisation and shrinkage methods widely used for subset selection and estimation in regression problems. From a Bayesian perspective, the Lasso-type estimate can be viewed as a Bayesian posterior mode when specifying independent Laplace prior distributions for the coefficients of independent variables [32 T. Park, G. Casella, The Bayesian Lasso, J. Amer. Statist. Assoc. 103 (2008), pp. 681–686. doi: 10.1198/016214508000000337[Taylor &; Francis Online], [Web of Science ®] , [Google Scholar]]. A scale mixture of normal priors can also provide an adaptive regularisation method and represents an alternative model to the Bayesian Lasso-type model. In this paper, we assign a normal prior with mean zero and unknown variance for each quantile coefficient of independent variable. Then, a simple Markov Chain Monte Carlo-based computation technique is developed for quantile regression (QReg) models, including continuous, binary and left-censored outcomes. Based on the proposed prior, we propose a criterion for model selection in QReg models. The proposed criterion can be applied to classical least-squares, classical QReg, classical Tobit QReg and many others. For example, the proposed criterion can be applied to rq(), lm() and crq() which is available in an R package called Brq. Through simulation studies and analysis of a prostate cancer data set, we assess the performance of the proposed methods. The simulation studies and the prostate cancer data set analysis confirm that our methods perform well, compared with other approaches.     

Designing stratified sampling in economic and business surveys

In most economic and business surveys, the target variables (e.g. turnover of enterprises, income of households, etc.) commonly resemble skewed distributions with many small and few large units. In such surveys, if a stratified sampling technique is used as a method of sampling and estimation, the convenient way of stratification such as the use of demographical variables (e.g. gender, socioeconomic class, geographical region, religion, ethnicity, etc.) or other natural criteria, which is widely practiced in economic surveys, may fail to form homogeneous strata and is not much useful in order to increase the precision of the estimates of variables of interest. In this paper, a stratified sampling design for economic surveys based on auxiliary information has been developed, which can be used for constructing optimum stratification and determining optimum sample allocation to maximize the precision in estimate.     

Pharmacotherapy for obesity in menopausal women

Weight gain, during and after the menopause is common. Contributing factors include ethnicity, reduced physical activity, reduced lean mass, reduced resting metabolic rate and treatment with certain drugs, e.g. steroids, insulin, glitazones. Excess body weight increases the risk of medical conditions including type 2 diabetes, hypertension, osteoarthritis, certain cancers and is associated with increased mortality. This review examines pharmacological approaches to promote weight loss. Pharmacological therapy should be considered as an adjunct to diet and lifestyle changes. The licensed drugs orlistat, sibutramine and rimonabant are discussed. Obesity increases the risk of type 2 diabetes. Thus, the effects of metformin and exenatide are examined.     

Myopic risk-seeking: The impact of narrow decision bracketing on lottery play

In two experiments conducted with low-income participants, we find that individuals are more likely to buy state lottery tickets when they make several purchase decisions one-at-a-time, i.e. myopically, than when they make one decision about how many tickets to purchase. These results extend earlier findings showing that “broad bracketing” of decisions encourages behavior consistent with expected value maximization. Additionally, the results suggest that the combination of myopic decision making and the “peanuts effect”—greater risk seeking for low stakes than high stakes gambles—can help explain the popularity of state lotteries.

The challenges and coping strategies of single mothers caring for children with special needs: Experiences from Kashmir,India

Families play a significant role in the development of their children. Nearly 16% of children in the industrialized world live in single-parent households, which are mostly comprised of single mothers. Although motherhood is the most beautiful feeling a woman cherishes, this experience can be challenging when the woman has to raise their special-needs child on her own, without the support of her spouse or partner. This study aims to understand the challenges and coping strategies of single mothers while raising their special-needs children. To this end, in-depth case interviews were conducted with three single mothers selected purposively from the Srinagar area in Kashmir, India. The results revealed that multiple roles and responsibilities had an adverse effect on the psychological well-being of single mothers who had to deal with the financial burden of caring for their child's condition, including expenses for doctor visits, special therapies and medication, among other. In addition, they had to experience the social pressure and stigmatization attached to being a single mother. The most widely employed coping strategies include support from family and friends, religious faith, and self-coping. The study recommends the creation of welfare packages like free counseling, stipend support, inclusive education, free medical treatment, and therapies for Children with Special Needs (CWSNs) and their mothers.     

Bayesian bridge quantile regression

Regularization methods for simultaneous variable selection and coefficient estimation have been shown to be effective in quantile regression in improving the prediction accuracy. In this article, we propose the Bayesian bridge for variable selection and coefficient estimation in quantile regression. A simple and efficient Gibbs sampling algorithm was developed for posterior inference using a scale mixture of uniform representation of the Bayesian bridge prior. This is the first work to discuss regularized quantile regression with the bridge penalty. Both simulated and real data examples show that the proposed method often outperforms quantile regression without regularization, lasso quantile regression, and Bayesian lasso quantile regression.     

Working with Stepfamilies

Stepfamilies are complex family systems that warrant a specific model to guide therapy practice. Once a stepfamily has formed it's easy to overlook the lack of an attachment history that is commonly embedded in a biologically connected nuclear family. This can result in stepparents picking up parental responsibility for their stepchildren, which often may not go well. This paper highlights the need for clarity concerning the different levels of connection within a stepfamily, and the importance of avoiding ‘nuclear family‐style’ solutions and assumed attachments. This is especially important in the early stages of the relationship when everyone is adjusting to changing circumstances, which is often a time when issues of loyalty and betrayal fuel many of the actions taken. A range of family therapy techniques can be helpfully adapted to working with stepfamilies, especially as relationships with children often bring them to therapy. In particular, therapists can utilise ideas from structural family therapy to help guide the stepfamily to navigate the complexities of everyday life.     

A GQL estimation approach for analysing non-stationary over-dispersed BINAR(1) time series

This paper proposes a generalized quasi-likelihood (GQL) function for estimating the vector of regression and over-dispersion effects for the respective series in the bivariate integer-valued autoregressive process of order 1 (BINAR(1)) with Negative Binomial (NB) marginals. The auto-covariance function in the proposed GQL is computed using some ‘robust’ working structures. As for the BINAR(1) process, the inter-relation between the series is induced mainly by the correlated NB innovations that are subject to different levels of over-dispersion. The performance of the GQL approach is tested via some Monte-Carlo simulations under different combination of over-dispersion together with low and high serial- and cross-correlation parameters. The model is also applied to analyse a real-life series of day and night accidents in Mauritius.