Childless Elders in Assisted Living: Findings from the Maryland Assisted Living Study

We compared data drawn from a random sample of 399 current assisted living residents and a subsample of 222 newly admitted residents for two groups: childless residents and residents with children. The percentage of childless residents (26%) in our study was slightly higher than U.S. population estimates of childless individuals aged 65 years and older (20%). In the overall sample, the two groups differed significantly by age, race, and women's years of education. The childless group was slightly younger, had a higher percentage of African American residents, and had more years of education than the group with children. In the subsample, we looked at demographic, functional, financial, and social characteristics and found that childless residents reported fewer diagnoses of dementia and fewer visits from a relative but more reported paying less money per month for assisted living and having private insurance than residents with children. As childlessness among older adults continues to increase, it will become increasingly important to understand how child status affects the need for and experience of long-term care.     

Maternal education and child health: Is there a strong causal relationship?

Using data from the first round of Demographic and Health Surveys for 22 developing countries, we examine the effect of maternal education on three markers of child health: infant mortality, children s height-for-age, and immunization status. In contrast to other studies, we argue that although there is a strong correlation between maternal education and markers of child health, a causal relationship is far from established. Education acts as a proxy for the socioeconomic status of the family and geographic area of residence. Introducing controls for husband’s education and access to piped water and toilet attenuate the impact of maternal education on infant mortality and children’s height-for-age. This effect is further reduced by controlling for area of residence through the use of fixed-effects models. In the final model. maternal education has a statistically significant impact on infant mortality and height-forage in only a handful of countries. In contrast. maternal education remains statistically significant for chidren’s immunization status in about one-half of the countries even after individual-level and community-level controls are introduced.     

Inference based on progressive Type I interval censored data from log-normal distribution

This article considers inference for the log-normal distribution based on progressive Type I interval censored data by both frequentist and Bayesian methods. First, the maximum likelihood estimates (MLEs) of the unknown model parameters are computed by expectation-maximization (EM) algorithm. The asymptotic standard errors (ASEs) of the MLEs are obtained by applying the missing information principle. Next, the Bayes’ estimates of the model parameters are obtained by Gibbs sampling method under both symmetric and asymmetric loss functions. The Gibbs sampling scheme is facilitated by adopting a similar data augmentation scheme as in EM algorithm. The performance of the MLEs and various Bayesian point estimates is judged via a simulation study. A real dataset is analyzed for the purpose of illustration.     

Bayesian Accelerated Life Testing under Competing Weibull Causes of Failure

Consider a J-component series system which is put on Accelerated Life Test (ALT) involving K stress variables. First, a general formulation of ALT is provided for log-location-scale family of distributions. A general stress translation function of location parameter of the component log-lifetime distribution is proposed which can accommodate standard ones like Arrhenius, power-rule, log-linear model, etc., as special cases. Later, the component lives are assumed to be independent Weibull random variables with a common shape parameter. A full Bayesian methodology is then developed by letting only the scale parameters of the Weibull component lives depend on the stress variables through the general stress translation function. Priors on all the parameters, namely the stress coefficients and the Weibull shape parameter, are assumed to be log-concave and independent of each other. This assumption is to facilitate Gibbs sampling from the joint posterior. The samples thus generated from the joint posterior is then used to obtain the Bayesian point and interval estimates of the system reliability at usage condition.     

Marine protected areas and children’s dietary diversity in the Philippines

Fish living around the coral reefs in the Philippines provide livelihoods for more than a million local fishers and are an important source of protein for coastal communities. However, this rich resource is at risk from myriad threats, which consequently threaten human livelihoods, nutrition, and health. In this paper, we examine the degree to which marine protected areas (MPAs), which aim to conserve marine biodiversity, are associated with improved nutritional outcomes in children under age 5. This analysis, which uses data from the 2008 Philippines Demographic and Health Survey and MPA data from the Coastal Conservation and Education Fund, found a positive association between MPAs and children’s dietary diversity when the MPAs were located closer than 2 km to a child’s community. MPA characteristics such as age or type of management were not consistently associated with dietary diversity. These results suggest a positive association of proximity to MPAs with certain aspects of children’s diet.     

A Generalization of Shapiro–Wilk's Test for Multivariate Normality

A goodness-of-fit test for multivariate normality is proposed which is based on Shapiro–Wilk's statistic for univariate normality and on an empirical standardization of the observations. The critical values can be approximated by using a transformation of the univariate standard normal distribution. A Monte Carlo study reveals that this test has a better power performance than some of the best known tests for multinormality against a wide range of alternatives.     

Identifiability and Comparison of Estimation Methods on Weibull Mixture Models

In this article, the finite mixture model of Weibull distributions is studied, the identifiability of the model with m components is proven, and the parameter estimators for the case of two components resulted by several algorithms are compared. The parameter estimators are obtained with maximum likelihood performing calculations with different algorithms: expectation-maximization (EM), Fisher scoring, backfitting, optimization of k-nearest neighbor approach, and random walk algorithm using Monte Carlo simulation. The Akaike information criterion and the log-likelihood value are used to compare models. In general, the proposed random walk algorithm shows better performance in mean square error and bias. Finally, the results are applied to electronic component lifetime data.     

Security Assurance: How Online Service Providers Can Influence Security Control Perceptions and Gain Trust

Security researchers agree that security control is a difficult to observe credence quality of online services that Internet users cannot easily assess through research or experience. Yet there is evidence that users form perceptions of security control that strongly determine how much trust they put in online services. This study investigates whether users’ security control perceptions arise solely from their predispositions or whether online service providers can influence them. The study also examines whether these seemingly undependable perceptions of security control lead to trust or whether more traditional factors might offer a better explanation of trust under security risks. To address these issues, this study proposes a new theory of security assurance that integrates the frameworks of trust and quality signals. The results show that rather than being guided by predispositions, users appear to mainly assess security control based on indirect cues controlled by service providers. Importantly, Internet users do not treat the credence quality of security the same way they treat qualities that can be understood through search and experience. Although returning users develop security control perceptions and trust from the usual heuristics of ongoing relationships, they also continue to evaluate market information about service providers like they do in new relationships. The proposed model offers a new perspective of how users respond to the uncertain and technically challenging qualities prevalent in online services.     

Probabilistic Modeling of Flood Hazard and its Risk Assessment for Eastern Region of India

This article models flood occurrence probabilistically and its risk assessment. It incorporates atmospheric parameters to forecast rainfall in an area. This measure of precipitation, together with river and ground parameters, serve as parameters in the model to predict runoff and subsequently inundation depth of an area. The inundation depth acts as a guide for predicting flood proneness and associated hazard. The vulnerability owing to flood has been analyzed as social vulnerability $(V_S)$, vulnerability to property $(V_P)$, and vulnerability to the location in terms of awareness $(V_A)$. The associated risk has been estimated for each area. The distribution of risk values can be used to classify every area into one of the six risk zones—namely, very low risk, low risk, moderately low risk, medium risk, high risk, and very high risk. The prioritization regarding preparedness, evacuation planning, or distribution of relief items should be guided by the range on the risk scale within which the area under study falls. The flood risk assessment model framework has been tested on a real‐life case study. The flood risk indices for each of the municipalities in the area under study have been calculated. The risk indices and hence the flood risk zone under which a municipality is expected to lie would alter every day. The appropriate authorities can then plan ahead in terms of preparedness to combat the impending flood situation in the most critical and vulnerable areas.     

Accuracy of Simulations for Stochastic Dynamic Models

This paper is concerned with accuracy properties of simulations of approximate solutions for stochastic dynamic models. Our analysis rests upon a continuity property of invariant distributions and a generalized law of large numbers. We then show that the statistics generated by any sufficiently good numerical approximation are arbitrarily close to the set of expected values of the model's invariant distributions. Also, under a contractivity condition on the dynamics, we establish error bounds. These results are of further interest for the comparative study of stationary solutions and the estimation of structural dynamic models.