Bayesian model selection for join point regression with application to age-adjusted cancer rates

Summary.  The method of Bayesian model selection for join point regression models is developed. Given a set of K +1 join point models M ₀,  M ₁, …,  M _K with 0, 1, …,  K join points respec-tively, the posterior distributions of the parameters and competing models M _k are computed by Markov chain Monte Carlo simulations. The Bayes information criterion BIC is used to select the model M _k with the smallest value of BIC as the best model. Another approach based on the Bayes factor selects the model M _k with the largest posterior probability as the best model when the prior distribution of M _k is discrete uniform. Both methods are applied to analyse the observed US cancer incidence rates for some selected cancer sites. The graphs of the join point models fitted to the data are produced by using the methods proposed and compared with the method of Kim and co-workers that is based on a series of permutation tests. The analyses show that the Bayes factor is sensitive to the prior specification of the variance σ ², and that the model which is selected by BIC fits the data as well as the model that is selected by the permutation test and has the advantage of producing the posterior distribution for the join points. The Bayesian join point model and model selection method that are presented here will be integrated in the National Cancer Institute's join point software ( http://www.srab.cancer.gov/joinpoint/ ) and will be available to the public.     

Estimation of a bivariate distribution function from incomplete observations

Kaplan and Meier (1958) give a maximum likelihood estimator of the distribution function based on a univariate right censored sample-Here we investigate the extension of their results to the case of bivariate right censored samples. Following Efron (1967), we provide "self-consistent" estimators for the bivariate distribution function.     

A unified approach to estimating parameters in some generalized poission distributions

A family of distributions labelled as Poisson v Katz is formulated, which includes, as particular or limiting cases, the Negative Binomial, Neyman Type A, Poisson v Pascal, and Poisson v Binomial. Thus, while analyzing data, estimating the parameters in the Poisson v Katz family obviates the necessity of having to choose from among the particular or limiting cases. In this article minimum chi-square estimators are presented and their asymptotic relative efficiency obtained. An example is presented to illustrate the procedure     

Inflated modified power series distributions with applications

In certain applications involving discrete data, it is sometimes found that X = 0 is observed with a frequency significantly higher than predicted by the assumed model. Zero inflated Poisson, binomial and negative binomial models have been employed in some clinical trials and in some regression analysis problems.

In this paper, we study the zero inflated modified power series distributions (IMPSD) which include among others the generalized Poisson and the generalized negative binomial distributions and hence the Poisson, binomial and negative binomial distributions. The structural properties along with the distribution of the sum of independent IMPSD variables are studied. The maximum likelihood estimation of the parameters of the model is examined and the variance-covariance matrix of the estimators is obtained. Finally, examples are presented for the generalized Poisson distribution to illustrate the results.     

Parameter estimation of nonlinear mixed-effects models using first-order conditional linearization and the EM algorithm

Nonlinear mixed-effects (NLME) models are flexible enough to handle repeated-measures data from various disciplines. In this article, we propose both maximum-likelihood and restricted maximum-likelihood estimations of NLME models using first-order conditional expansion (FOCE) and the expectation–maximization (EM) algorithm. The FOCE-EM algorithm implemented in the ForStat procedure SNLME is compared with the Lindstrom and Bates (LB) algorithm implemented in both the SAS macro NLINMIX and the S-Plus/R function nlme in terms of computational efficiency and statistical properties. Two realworld data sets an orange tree data set and a Chinese fir (Cunninghamia lanceolata) data set, and a simulated data set were used for evaluation. FOCE-EM converged for all mixed models derived from the base model in the two realworld cases, while LB did not, especially for the models in which random effects are simultaneously considered in several parameters to account for between-subject variation. However, both algorithms had identical estimated parameters and fit statistics for the converged models. We therefore recommend using FOCE-EM in NLME models, particularly when convergence is a concern in model selection.     

The role of human and social board capital in driving CSR reporting

The objective of this paper is to analyze the effect of the professional, technical and relational background (human and social capital) of outside directors on promoting firm CSR disclosure. Following the Hillman et al. (2000) taxonomy of board members, we classify outside directors as business experts, support specialists and community influential, and examine whether business and technical expertise or political ties in the boardroom affect CSR disclosure.This study confirms that not all outside directors are equally effective in improving CSR disclosure and that only certain kinds of outside directors, those classified as support specialists, help promote it. On the other hand, our findings also show that directors with previous experience as politicians affect CSR disclosure negatively, probably due to their interests in safeguarding their reputation within the company, in avoiding public scrutiny and in protecting their political connections. In addition, our set of analysis with interaction effects reveals that powerful CEOs have the incentive to promote CSR-related strategies and to convince business experts and support specialist directors to enhance profitable sustainability strategies and transparency in CSR disclosure. Nevertheless, the powerful CEO effect is not enough to compensate the negative role of political directors on CSR reporting. Therefore, this paper supports the theories in favor of analyzing the multiple configurations of corporate governance mechanisms by adopting a holistic approach, and the need to combine these configurations in order to analyze their impact on CSR behavior.     

Are college students health literate?

Abstract

Objective: To examine the health literacy of college students. Participants: A convenience sample of 245 graduate and undergraduate college students. Methods: During February–April of 2018 participants completed the Short Test of Functional Health Literacy which assessed literacy on two passages describing a thyroid scan, and basic healthcare insurance information. Results: Most college students displayed adequate health literacy (99.2%). The ANOVA analyses revealed college classification was the most significant predictor, followed by ethnicity, sex, and primary language. Age was significantly related to health literacy, when holding college classification constant. Interestingly, college major, healthcare work experience, or having health care credentials were not predictors of health literacy. Conclusion: This group of college students had adequate health literacy skills. However, the results of this study revealed demographic disparities that suggest further study.     

Acquisition of Telecommunications Bandwidth Under Economies of Scale in Size and Duration of Contracts*

Significant advances in information technology have brought about increased demand for bandwidth. Buyers of bandwidth often encounter bandwidth prices that are decreasing over time. Additionally, bandwidth prices at any point in time are decreasing in total bandwidth purchased and length of contracts. Therefore, buyers face complex decisions in terms of the number of contracts to buy, their bandwidth, and their lengths. In this article, we formulate models for the acquisition of bandwidth from a buyer's perspective. We begin with a model that allows varying contract durations under deterministic demand and without allowing shortages or overlapping contracts. We then formulate a simpler model, which restricts contract lengths over the planning horizon to be equal. We also solve the problem under probabilistic demand and allowing for shortages, which are satisfied by buying additional bandwidth at a premium. We perform numerical sensitivity analysis to compare the results of the models and illustrate the results with numerical examples. The numerical analyses illustrate that using relatively simple equal‐length contracts produces approximately the same results as the more complicated unequal‐length contract strategy.     

Causal Linkages in Supply Chain Management: An Exploratory Study of North American Manufacturing Firms

In this paper we investigate the key causal linkages in supply chain management. We propose a conceptual framework and test this framework on data from 215 North American manufacturing firms using structural equation modeling techniques. Three major research issues are addressed in this study: Do sourcing decisions affect the degree to which firms achieve manufacturing goals of cost, flexibility, dependability, and quality? Does the degree of manufacturing goal achievement lead to higher customer responsiveness? Does the degree of manufacturing goal achievement lead to higher internal manufacturing performance? The study examines the relationship among sourcing decisions, manufacturing goals, customer responsiveness, and manufacturing performance. The results support the notion that an integrated supply chain involves aligning sourcing decisions to achieve manufacturing goals that are set to respond favorably to the needs of customers.     

World Income Distribution and Tax Reform: What Tax Systems Do Low‐Income Countries Need?

This article develops a new method for assessing relative direct tax burdens across all countries, treating the world as a single economic entity and assuming identical preferences across countries. Empirical results show that the new direct tax burden indices are significantly high in low‐income countries in comparison with middle‐ and high‐income countries. This article argues in favour of narrowing the base of income and capital gains tax in low‐income countries and a long‐term convergence of the tax burden levels across countries. Future research into tax reforms in low‐income countries should focus simultaneously on economic growth, quality of life and the natural environment.