Artem Prokhorov

Elder Abuse in Long-Term Care: Types, Patterns, and Risk Factors

The authors investigated types and patterns of elder abuse by paid caregivers in long-term care and assessed the role of several risk factors for different abuses and for multiple abuse types. The results are based on a 2005 random-digit-dial survey of relatives of persons in long-term care. We computed occurrence rates and conditional occurrence rates for each of six abuse types: physical, caretaking, verbal, emotional, neglect, and material. Among older adults who have experienced at least one type of abuse, more than half (51.4%) have experienced another type of abuse. Physical functioning problems, activities of daily living limitations, and behavioral problems are significant risk factors for at least three types of abuse and are significant for multiple abuse types. The findings have implications for those monitoring the well-being of older adults in long-term care as well as those responsible for developing public health interventions.

The Effect of Care Setting on Elder Abuse: Results from a Michigan Survey

This study compares abuse rates for elders age 60 and older in three care settings: nursing home, paid home care, and assisted living. The results are based on a 2005 random-digit dial survey of relatives of, or those responsible for, a person in long-term care. Nursing homes have the highest rates of all types of abuse, although paid home care has a relatively high rate of verbal abuse and assisted living has an unexpected high rate of neglect. Even when adjusting for health conditions, care setting is a significant factor in both caretaking and neglect abuses. Moving from paid home care to nursing homes is shown to more than triple the odds of neglect. Furthermore, when computing abuse rates by care setting for persons with specified health conditions, nursing homes no longer have the highest abuse rates.

Using copulas to model time dependence in stochastic frontier models

We consider stochastic frontier models in a panel data setting where there is dependence over time. Current methods of modeling time dependence in this setting are either unduly restrictive or computationally infeasible. Some impose restrictive assumptions on the nature of dependence such as the “scaling” property. Others involve T-dimensional integration, where T is the number of cross-sections, which may be large. Moreover, no known multivariate distribution has the property of having commonly used, convenient marginals such as normal/half-normal. We show how to use copulas to resolve these issues. The range of dependence we allow for is unrestricted and the computational task involved is easy compared to the alternatives. Also, the resulting estimators are more efficient than those that assume independence over time. We propose two alternative specifications. One applies a copula function to the distribution of the composed error term. This permits the use of maximum likelyhood estimate (MLE) and generalized method moments (GMM). The other applies a copula to the distribution of the one-sided error term. This allows for a simulated MLE and improved estimation of inefficiencies. An application demonstrates the usefulness of our approach.

Impairment and Abuse of Elderly by Staff in Long-Term Care in Michigan: Evidence From Structural Equation Modeling:

Elder abuse in long-term care has become a very important public health concern. Recent estimates of elder abuse prevalence are in the range of 2% to 10% (Lachs & Pillemer, 2004), and current changes in population structure indicate a potential for an upward trend in prevalence (Malley-Morrison, Nolido, & Chawla, 2006; Post et al., 2006). More than 20 years ago, Karl Pillemer called for sociological research on patient maltreatment in nursing homes and provided an overview model for the conduct of such research (Pillemer, 1988). The research literature since then has not provided the definitive model to account for patient maltreatment that Pillemer hoped for. Instead, it has produced a laundry list of risk factors that includes the patient’s functional disability, cognitive impairment, social isolation, age, race, income, family background, life events, dementia, and depression (Dyer, Pavlik, Murphy, & Hyman, 2000; Lachs & Pillemer, 2004; Lachs,Williams, Obrien, Hurst, & Horwitz, 1997; Pavlik, Hyman, Festa, & Dyer, 2001; Schofield & Mishra, 2003). However, no theory exists to place these factors in a causal structure that relates the factors to each other and to whether abuse occurs. This study is a first step in that direction. Nine hypotheses were generated focusing on the effects of two dimensions of impairment—(a) physical and cognitive and (b) age and behavior problems—on susceptibility to abuse among elderly in long-term care.The relationships between factors and from factors to susceptibility to abuse are specified in a structural equation model where “susceptibility to abuse,” “physical impairment,” and “cognitive impairment” are latent variables, and behavior problems and age are directly measured.

Heavy Tails and Copulas:Topics in Dependence Modelling in Economics and Finance

This book offers a unified approach to the study of crises, large fluctuations, dependence and contagion effects in economics and finance. It covers important topics in statistical modeling and estimation, which combine the notions of copulas and heavy tails — two particularly valuable tools of todays research in economics, finance, econometrics and other fields — in order to provide a new way of thinking about such vital problems as diversification of risk and propagation of crises through financial markets due to contagion phenomena, among others. The aim is to arm todays economists with a toolbox suited for analyzing multivariate data with many outliers and with arbitrary dependence patterns. The methods and topics discussed and used in the book include, in particular, majorization theory, heavy-tailed distributions and copula functions — all applied to study robustness of economic, financial and statistical models, and estimation methods to heavy tails and dependence.

Copula based factorization in Bayesian multivariate infinite mixture models

Bayesian nonparametric models based on infinite mixtures of density kernels have been recently gaining in popularity due to their flexibility and feasibility of implementation even in complicated modeling scenarios. However, these models have been rarely applied in more than one dimension. Indeed, implementation in the multivariate case is inherently difficult due to the rapidly increasing number of parameters needed to characterize the joint dependence structure accurately. In this paper, we propose a factorization scheme of multivariate dependence structures based on the copula modeling framework, whereby each marginal dimension in the mixing parameter space is modeled separately and the marginals are then linked by a nonparametric random copula function. Specifically, we consider nonparametric univariate Gaussian mixtures for the marginals and a multivariate random Bernstein polynomial copula for the link function, under the Dirichlet process prior. We show that in a multivariate setting this scheme leads to an improvement in the precision of a density estimate relative to the commonly used multivariate Gaussian mixture. We derive weak posterior consistency of the copula-based mixing scheme for general kernel types under high-level conditions, and strong posterior consistency for the specific Bernstein–Gaussian mixture model.

Aging and elder abuse: Projections for Michigan

Lori Post1, Charles Salmon1, Artem Prokhorov2, James Oehmke3 and Sarah Swierenga4 1College of Communication Arts & Sciences, Michigan State University, East Lansing, MI, USA 2Department of Economics, Concordia University, Montreal, Quebec, Canada 3Agricultural, Food & Resource Economics, Michigan State University, East Lansing, MI, USA 4Office of Accessibility & Usability, Michigan State University, East Lansing, MI, USA

Two‐sample nonparametric estimation of intergenerational income mobility in the United States and Sweden

We estimate intergenerational income mobility in the US and Sweden, using a new nonparametric approach. The approach addresses several empirical issues raised in the literature and applies when other estimators are infeasible. We argue that previous estimates of income mobility conceal the heterogeneous nature of the transmission mechanism by keeping mobility constant across families. The striking differences we find between mobility patterns across family backgrounds, captured by fathers education, lead us to question the conventional result that intergenerational transmission of earnings is weaker in Sweden than in the United States, for important parts of the population.

Bartlett-type Correction of Distance Metric Test

We derive a corrected distance metric (DM) test of general restrictions. The correction factor depends on the value of the uncorrected statistic and the new statistic is Bartlett-type. In the setting of covariance structure models, we show using simulations that the quality of the new approximation is good and often remarkably good. Especially at around the 95th percentile, the distribution of the corrected test statistic is strikingly close to the relevant asymptotic distribution. This is true for various sample sizes, distributions, and degrees of freedom of the model. As a by-product we provide an intuition for the well-known observation in labor economic applications that using longer panels results in a reversal of the original inference.

The ε-complexity of copulas

We introduce the concept of ε-complexity of copula functions, discuss its properties and show how to use it in assessing copula specifications based on empirical copulas. The ε-complexity of copulas can be captured by two parameters of a linear regression which are easy to estimate and test. We provide an application focusing on bivariate modeling of stock returns.