On population-based simulation for static inference

In this paper we present a review of population-based simulation for static inference problems. Such methods can be described as generating a collection of random variables {X _n } _n=1,…,N in parallel in order to simulate from some target density π (or potentially sequence of target densities). Population-based simulation is important as many challenging sampling problems in applied statistics cannot be dealt with successfully by conventional Markov chain Monte Carlo (MCMC) methods. We summarize population-based MCMC (Geyer, Computing Science and Statistics: The 23rd Symposium on the Interface, pp. 156–163, 1991; Liang and Wong, J. Am. Stat. Assoc. 96, 653–666, 2001) and sequential Monte Carlo samplers (SMC) (Del Moral, Doucet and Jasra, J. Roy. Stat. Soc. Ser. B 68, 411–436, 2006a), providing a comparison of the approaches. We give numerical examples from Bayesian mixture modelling (Richardson and Green, J. Roy. Stat. Soc. Ser. B 59, 731–792, 1997).     

STATISTICAL INFERENCE OF TRANSMISSION FIDELITY OF DNA METHYLATION PATTERNS OVER SOMATIC CELL DIVISIONS IN MAMMALS

We develop Bayesian inference methods for a recently-emerging type of epigenetic data to study the transmission fidelity of DNA methylation patterns over cell divisions. The data consist of parent-daughter double-stranded DNA methylation patterns with each pattern coming from a single cell and represented as an unordered pair of binary strings. The data are technically difficult and time-consuming to collect, putting a premium on an efficient inference method. Our aim is to estimate rates for the maintenance and de novo methylation events that gave rise to the observed patterns, while accounting for measurement error. We model data at multiple sites jointly, thus using whole-strand information, and considerably reduce confounding between parameters. We also adopt a hierarchical structure that allows for variation in rates across sites without an explosion in the effective number of parameters. Our context-specific priors capture the expected stationarity, or near-stationarity, of the stochastic process that generated the data analyzed here. This expected stationarity is shown to greatly increase the precision of the estimation. Applying our model to a data set collected at the human FMR1 locus, we find that measurement errors, generally ignored in similar studies, occur at a non-trivial rate (inappropriate bisulfite conversion error: 1.6% with 80% CI: 0.9-2.3%). Accounting for these errors has a substantial impact on estimates of key biological parameters. The estimated average failure of maintenance rate and daughter de novo rate decline from 0.04 to 0.024 and from 0.14 to 0.07, respectively, when errors are accounted for. Our results also provide evidence that de novo events may occur on both parent and daughter strands: the median parent and daughter de novo rates are 0.08 (80% CI: 0.04-0.13) and 0.07 (80% CI: 0.04-0.11), respectively.     

Political Corporate Social Responsibility: Reviewing Theories and Setting New Agendas

There has been rising interest in political corporate social responsibility (political CSR), defined as activities where CSR has an intended or unintended political impact, or where intended or unintended political impacts on CSR exist. Based on a survey and content analysis of 146 peer‐reviewed academic articles from 18 journals over the 14‐year period 2000–2013, this paper systematically reviews the existing applications of general theories (such as legitimacy theory, the resource‐based view and Habermasian political theory) within the political CSR literature. The survey indicates that the political CSR field is dominated by institutional theory and stakeholder theory, but future theory development needs to go beyond these theories in order to address a number of critical gaps. This review specifically points to several avenues for future political CSR research with regard to the individual level of analysis, domain integration and political CSR in multinational enterprises. The paper ends with a call for a new theory‐informed and pluralist research agenda on political CSR to integrate different perspectives and re‐examine the role of the state.     

On inference from Markov chain macro-data using transforms

We consider data that are longitudinal, arising from n individuals over m time periods. Each individual moves according to the same homogeneous Markov chain, with s states. If the individual sample paths are observed, so that ‘micro-data’ are available, the transition probability matrix is estimated by maximum likelihood straightforwardly from the transition counts. If only the overall numbers in the various states at each time point are observed, we have ‘macro-data’, and the likelihood function is difficult to compute. In that case a variety of methods has been proposed in the literature. In this paper we propose methods based on generating functions and investigate their performance.     

The Value of Children,Parenting Strains,and Depression Among Urban African American Mothers

This exploratory analysis examines the relationship between two dimensions of the parenting experience—values regarding children and parenting strains—and depression in a sample of 285 urban, economically disadvantaged African American women. An examination of this relationship among economically disadvantaged African American women is important in light of the high stress exposure of these mothers, their limited opportunity to engage in rewarding employment, and in view of the centrality of motherhood in this cultural context. Our central finding is that the impact of parenting strains on depression is substantially moderated by beliefs about the costs and emotional fulfillment associated with having children. African American mothers who regard the costs of children as low and the potential of children for providing emotional fulfillment as high are significantly more depressed by high parenting strains than African American mothers who hold the opposite beliefs about the costs and benefits of children. Variation in values regarding children and parenting strains accounts for more than 40% of variation in depression. These findings suggest the importance of understanding the social–psychological factors that contribute to the quality of parenting, not only because maternal mental health is important in and of itself, but because maternal well-being has a direct impact on child well-being.     

Mental Illness and the Criminal Justice System

This is the second essay in a two‐part series exploring the relationships between mental illness, criminal behavior, and the criminal justice system. The number of mentally ill persons in prisons and jails has increased substantially over the last several decades, and there are currently more people with mental illness behind bars than there are in mental hospitals. In this essay, we place these trends within a broader historical context of the social control of mental illness in the United States. We identify how and why mentally ill persons have come to be overrepresented in the criminal justice system and highlight the unique challenges this population poses for police, courts, and correctional facilities. Finally, we review several recent innovations in policy and practice that may help alleviate that burden of criminal justice involvement on mentally ill offenders, as well as the burden of mentally ill offenders on the criminal justice system.     

The effect of local livestock population changes on auction market viability—a spatial analysis

This paper examines the relationship between auction market closures over the period 1980–2000 and livestock population changes, as recorded in the agricultural and horticultural census. Auction market locations and census data were collated within a Geographical Information System and changes in livestock populations examined by region and by market catchment. Regionally, auction market closures during the 1980s were significantly associated with concurrent reductions in cattle numbers, with market reductions following loss of cattle in eastern lowland areas. No such association with livestock numbers was found during the 1990s. Within these agricultural regions, individual market closures were not associated with changes in local livestock numbers within their catchments. Thus, the historical evidence suggests that whilst a substantial loss of livestock within a region puts pressure on its network of auction markets, the viability of individual markets within the region is decided by other factors. Such factors include the capital costs of modernising market facilities and meeting new regulatory requirements, the effect of unpaid debt on market cashflow, and the market's ability to diversify into other areas of business. The recent outbreak of foot and mouth disease may lead to further market closures, both directly as a result of livestock culls and indirectly as farmers continue to use alternative marketing channels developed when auction markets were closed during the outbreak.     

Occupational overuse syndrome and the effects of psychosocial stressors on keyboard users in the newspaper industry

There is continuing concern about the incidence of occupational overuse syndrome (OOS) among workers using keyboards. Following results from research showing positive correlations between psychosocial factors in the workplace and the incidence of OOS symptoms, it was hypothesizcd that differences in rates of pain reporting by keyboard workers are related to job stress caused by different levels in thc quality of the psychosocial work environment. An interaction between autonomy and the impact of work pressure on pain reporting was also predicted. Data were used from a sample of 550 keyboard users in several different types of job, working in eight different newspaper offices of the same newspaper company. The results confirmed the hypotheses. Post hoc analysis showed that there were important differences between the types of stressors that predicted the reporting of pain between offices, and those that predicted pain reporting between types of job. These results confirm the importance of considering psychosocial factors in work and workplace design in connection with the prevention of OOS. Conclusions are also drawn about the situation specificity of such empirical research, and the need for more theoretical work in the search for the aetiology of OOS.