Secure Bayesian model averaging for horizontally partitioned data

When multiple data owners possess records on different subjects with the same set of attributes—known as horizontally partitioned data—the data owners can improve analyses by concatenating their databases. However, concatenation of data may be infeasible because of confidentiality concerns. In such settings, the data owners can use secure computation techniques to obtain the results of certain analyses on the integrated database without sharing individual records. We present secure computation protocols for Bayesian model averaging and model selection for both linear regression and probit regression. Using simulations based on genuine data, we illustrate the approach for probit regression, and show that it can provide reasonable model selection outputs.     

Two tests of the erogodicity of monthly security return distributions

A popular technique in finance research is to use time averaged statistics from monthly security return time series as estimates of the properties of the process generating the returns. This procedure requires the process generating returns to be ergodic. Two tests of this property are applied to monthly return data on 160 securities. The trends tests on mean and variance supports the ergodicity property on two-thirds of the securities. The rank sum test supports this property on less than half the tests. The conclusion is against the ergodicity assumption.     

College students and tattoos. Influence of image,identity, family,and friends

Tattoos are popular among many groups of people. College students from a large university in the southwestern United States were queried using the 134-item Armstrong Team Tattoo Attitude Survey. Findings documents a progressive increase in tattooing and reflect significant support for tattoos, even from non-tattooed students. Similar and positive demographic characteristics were present in individuals who were both non-tattooed and tattooed. In addition, tattooed people were viewed positively, even among non-tattooed respondents. Friends significantly affected procurement and possession of tattoos, whereas of all family members, the sister was the only significant influence. Overall, friends, identity, and image seem to be the major influences for obtaining tattoos.     

Self-Rated Health and Attitudes About U.S. Health Care Policy

How does one’s own health influence attitudes about national health care policy? Previous research has documented a reliable (though bounded) role of personal interest in shaping policy preferences, but little is known about the role that personal health plays in shaping these attitudes. Using recent General Social Survey data, we find that those with poorer self-rated health were more likely to endorse a strong role for government in providing health care, and more likely to endorse increased spending on health care. Further analysis suggested that this relationship was moderated by political ideology. Implications for better understanding the source of health policy support are discussed.     

An Empirical Comparison of Multiple Imputation Methods for Categorical Data

Multiple imputation is a common approach for dealing with missing values in statistical databases. The imputer fills in missing values with draws from predictive models estimated from the observed data, resulting in multiple, completed versions of the database. Researchers have developed a variety of default routines to implement multiple imputation; however, there has been limited research comparing the performance of these methods, particularly for categorical data. We use simulation studies to compare repeated sampling properties of three default multiple imputation methods for categorical data, including chained equations using generalized linear models, chained equations using classification and regression trees, and a fully Bayesian joint distribution based on Dirichlet process mixture models. We base the simulations on categorical data from the American Community Survey. In the circumstances of this study, the results suggest that default chained equations approaches based on generalized linear models are dominated by the default regression tree and Bayesian mixture model approaches. They also suggest competing advantages for the regression tree and Bayesian mixture model approaches, making both reasonable default engines for multiple imputation of categorical data. Supplementary material for this article is available online.     

Multiple imputation for combining confidential data owned by two agencies

Summary.  Statistical agencies that own different databases on overlapping subjects can benefit greatly from combining their data. These benefits are passed on to secondary data analysts when the combined data are disseminated to the public. Sometimes combining data across agencies or sharing these data with the public is not possible: one or both of these actions may break promises of confidentiality that have been given to data subjects. We describe an approach that is based on two stages of multiple imputation that facilitates data sharing and dissemination under restrictions of confidentiality. We present new inferential methods that properly account for the uncertainty that is caused by the two stages of imputation. We illustrate the approach by using artificial and genuine data.     

Comparative use of complementary medicine, allied health, and manual therapies by middle-aged and older Australian women

We compared use of Complementary and Alternative Medicines (CAM) and therapies, Allied Health interventions (AH), and Manual Therapies (MT) in middle-aged and older Australian women. Cross-sectional data from the 2007 phase of the Longitudinal study of Aging in Women (LAW study) was analyzed. Self-reported 12-month usage patterns of CAM, AH, and MT were determined by mailed questionnaire. Results revealed that 56.2% of the middle-aged group, and 55.0% of the older group used CAM, AH, and MT services in the previous 12 months (a nonsignificant difference). In contrast, there was a highly significant difference between the overall use of CAM products by middle-aged women (88.2%) and older women (67.7%: p = .002).     

DSM-5: An Overview of Changes and Controversies

The DSM-5 offers many changes in the criteria and categories used in clinical diagnosis. The provocative and sometimes controversial nature of the changes has enlivened debate in the mental health field about how we should best understand our clients. I selectively survey what is new in DSM-5, why changes were made, and what about them is so controversial. First, I summarize the main metastructural and organizational changes, including elimination of the multiaxial system and changed chapter groupings. Second, I survey the most important new categories of disorder and the most important changes to the diagnostic criteria for existing categories of disorder. I focus on particularly controversial changes, such as those to substance use and addictive disorders, autism spectrum disorders, and posttraumatic stress disorder. Pros and cons are provided for changes in criteria as well as for the addition of new disorder categories, such as hoarding disorder and binge eating disorder. Finally, I offer a more in-depth review and analysis of the changes to the depressive disorders chapter, which was subject to some of the most intense controversies and had some of the most extensive changes.