MINING ERROR VARIANCE AND HITTING PAY-DIRT:

According to traditional error theory, sentiment measurements vary unsystematically from individual to individual. However, we find some patterned deviation in sentiments that characterize subsets of respondents within a seemingly homogeneous population. After demonstrating the existence of such patterns, we report an exploratory study aimed at identifying social characteristics of people with different patterns of sentiments. People embedded in multiple social networks have exaggerated sentiments that contour cultural patterns, and people with few social associations have attenuated sentiments conveying cultural detachment. In addition, people with disciplinary parents and romantic privation have some distinctive sentiments, depending on their gender.     

Estimation of noncentrality parameters

We propose some estimators of noncentrality parameters which improve upon usual unbiased estimators under quadratic loss. The distributions we consider are the noncentral chi-square and the noncentral F. However, we give more general results for the family of elliptically contoured distributions and propose a robust dominating estimator.     

The Micro System of Deviance in Intimate Relationships*

Deviance in intimate relationships may lie as much within the micro system of such relationships as it does within the personalities of the actors involved. Using a Parsonsian framework, it is argued that the strain of systemically disequilibrated relationships can produce deviant personality adaptations which, in turn, intensify conflict and deviance in those relationships. It is also proposed that a reciprocal quality exists to such maladaptations, with each member of the dyad contributing to the amplification of a stable system of deviant behavior patterns within the relationship. Because the patterns of maladaptive dyads spill over into parent-child relationships, they have implications for understanding intergenerational transmission of deviance between intimates.     

Effects of Medicaid managed care policies on mental health service use among a national probability sample of children in the child welfare system

Children in the child welfare system are dependent upon Medicaid to finance services for their considerable mental health needs. This study examines the effects of Medicaid policies on mental health service use among a national probability sample of children in the child welfare system. Data for this study came from the National Survey of Child and Adolescent Well-Being, the Caring for Children in Child Welfare study, and the Area Resource File. Weighted multivariate logistic regression analyses were conducted to estimate effects of policy variables on children's use of mental health services, controlling for child-level covariates and county-level health resources. Children in counties with behavioral carve-outs under Medicaid managed care had lower odds of inpatient mental health service use. Medicaid managed care enrollment and variations in type of provider reimbursement did not affect use of mental health services. Older age, greater need for mental health services, and higher levels of caregiver education were associated with increased odds of service use. Restrictions on use of inpatient mental healthcare caused by behavioral carve-outs may disproportionately affect children in the child welfare system who have high rates of such use. Careful adoption of carve-outs is necessary to assure appropriate care for these children.     

The Impact of Items and Anchors on Parent–Child Reports of Problem Behavior

Research has repeatedly demonstrated that parent and child reports of child behavior and emotional functioning often do not agree in terms of symptom severity or even symptom presence. Given the potential clinical impact that discrepant reports may have on the therapeutic process, a significant amount of research has addressed the factors influencing this observed discrepancy. Traditionally, this research has evaluated the impact of demographic factors on mean differences and obtained correlation coefficients between parent and child reports. The current research improves upon previous research in this area in two ways. First, the current research uses parent and child measures with identical rather than similar items as is commonly found in previous research. Second, the current research moves beyond identifying demographic influences on discrepant reports by explaining observed discrepancies in terms of differences in parent and child perceptions of typical child behavior.David Carlston and Benjamin Ogles are affiliated with Department of Psychology, Ohio University, Athens, USA.Funding for this project was provided by the Ohio Department of Mental Health.     

BAYESIAN SUBSET SELECTION AND MODEL AVERAGING USING A CENTRED AND DISPERSED PRIOR FOR THE ERROR VARIANCE

This article proposes a new data‐based prior distribution for the error variance in a Gaussian linear regression model, when the model is used for Bayesian variable selection and model averaging. For a given subset of variables in the model, this prior has a mode that is an unbiased estimator of the error variance but is suitably dispersed to make it uninformative relative to the marginal likelihood. The advantage of this empirical Bayes prior for the error variance is that it is centred and dispersed sensibly and avoids the arbitrary specification of hyperparameters. The performance of the new prior is compared to that of a prior proposed previously in the literature using several simulated examples and two loss functions. For each example our paper also reports results for the model that orthogonalizes the predictor variables before performing subset selection. A real example is also investigated. The empirical results suggest that for both the simulated and real data, the performance of the estimators based on the prior proposed in our article compares favourably with that of a prior used previously in the literature.     

Generalized additive models for location, scale and shape

Summary.  A general class of statistical models for a univariate response variable is presented which we call the generalized additive model for location, scale and shape (GAMLSS). The model assumes independent observations of the response variable y given the parameters, the explanatory variables and the values of the random effects. The distribution for the response variable in the GAMLSS can be selected from a very general family of distributions including highly skew or kurtotic continuous and discrete distributions. The systematic part of the model is expanded to allow modelling not only of the mean (or location) but also of the other parameters of the distribution of y , as parametric and/or additive nonparametric (smooth) functions of explanatory variables and/or random-effects terms. Maximum (penalized) likelihood estimation is used to fit the (non)parametric models. A Newton–Raphson or Fisher scoring algorithm is used to maximize the (penalized) likelihood. The additive terms in the model are fitted by using a backfitting algorithm. Censored data are easily incorporated into the framework. Five data sets from different fields of application are analysed to emphasize the generality of the GAMLSS class of models.