Upper bounds for the variances of certain random variables

Jacobson (1969) proved that the variance of a unimodal distribution on the unit interval is bounded above by 1/9. This paper generalizes Jacobson's result by deriving sharp upper bounds for the variances of α-unimodal distributions and for the variances of S-shaped distributions.     

A note on interpenetrating subsamples of unequal sizes drawn with and without replacement

In this article it is shown how one can gain in efficiency in selecting K interpenetrating subsamples of unequal sizes according to Chaudhuri and Adhikary's (1987) scheme rather than selecting them with replacement and considering the estimator based on distinct units as proposed by Bedi(1987).     

A two-stage unrelated randomized response model for estimating a rare sensitive attribute in probability proportional to size sampling using Poisson distribution

This paper describes the estimating procedures of mean number of entities that possess a rare sensitive attribute using the Mangat (1992) randomized device, when the population consists of some clusters and the population is again stratified with some clusters in each stratum. Unbiased estimation procedures for the mean number of individuals have been discussed and their properties are described when the parameter of a rare unrelated attribute is assumed to be known and unknown. An empirical study is carried out to show the dominance of the proposed estimator over Lee et al. (2013) estimator.     

Bayesian neural tree models for nonparametric regression

Frequentist and Bayesian methods differ in many aspects but share some basic optimal properties. In real-life prediction problems, situations exist in which a model based on one of the above paradigms is preferable depending on some subjective criteria. Nonparametric classification and regression techniques, such as decision trees and neural networks, have both frequentist (classification and regression trees (CARTs) and artificial neural networks) as well as Bayesian counterparts (Bayesian CART and Bayesian neural networks) to learning from data. In this paper, we present two hybrid models combining the Bayesian and frequentist versions of CART and neural networks, which we call the Bayesian neural tree (BNT) models. BNT models can simultaneously perform feature selection and prediction, are highly flexible, and generalise well in settings with limited training observations. We study the statistical consistency of the proposed approaches and derive the optimal value of a vital model parameter. The excellent performance of the newly proposed BNT models is shown using simulation studies. We also provide some illustrative examples using a wide variety of standard regression datasets from a public available machine learning repository to show the superiority of the proposed models in comparison to popularly used Bayesian CART and Bayesian neural network models.     

Projecting populations for major Pacific Island countries with and without COVID-19: pro-active insights for population policy

Journal of Population Research - The paper projects aggregate populations of six Pacific Island countries in both pre- and post-COVID19 scenarios using a Cohort Component Method for the period...     

Exploring the Rural-Agrarian Linkages of Human Trafficking: A Study of the Indian Punjab

This article explores the rural-agrarian linkages of human trafficking in the Indian Punjab. The study argues that ongoing agrarian crises, high risk-taking ability of some agrarian castes and low level of education in rural areas are fuelling the illicit business of human trafficking of Punjabis to foreign green pastures. Rural-agrarian communities are the main victims of the traffickers' exploitation. The study has wider policy implications as it suggests policy-makers should formulate a comprehensive policy framework for rural areas of Punjab to ensure the “3P” paradigm – prevention, protection and prosecution – of human trafficking.     

A multidimensional Lorenz dominance relation

This paper seeks to extend the unidimensional notion of Lorenz dominance to the multidimensional context. It formulates a definition of a multidimensional Lorenz dominance relation (MLDR) on the set of alternative distributions of well-being in an economy by incorporating a generalization of the well-known transfer principle of unidimensional theory suggested in recent literature. It also proposes two conditions which an MLDR may reasonably be required to satisfy. The paper notes that the existing literature does not seem to contain an example of an MLDR satisfying these two conditions and suggests one that does. The suggested MLDR does not seem to have appeared in the literature before.     

Focusing on Disability and Access in the Built Environment

This paper provides a documentation and discussion of the diverse experiences that different disabled people have with regards to access in the built environment. It begins by outlining the various ways in which disabled people's access needs and requirements are articulated in public policies and practices towards the development and regulation of the built environment. As the material indicates, disabled people's needs are poorly articulated and/or represented in the design and development of the built environment while the regulatory controls which oversee disabled people's access are weak. In the second part of the paper, disabled people's values, attitudes and practices towards access in the built environment are discussed by referring to the findings of focus group research. The material shows that many disabled people feel estranged and oppressed by facets of the built environment and generally feel powerless to do anything about it. We conclude by suggesting a number of ways of interconnecting the design and implementation of public policy towards the built environment with the daily lived experiences of disabled people.