Modified pseudo-likelihood estimator of the overdispersion parameter in Poisson mixture models

There are a variety of methods available for analyzing count data exhibiting extra-Poisson variation. Those based on quasi-likelihood, in particular, have gained tremendous popularity. In this paper, several moment estimators are compared. A new estimator, based on a modification of the pseudo-likelihood approach, is introduced and shown to perform very well when the overdispersion is not large, even if the sample size is very small.     

Prediction and Prevention of Child Abuse and Neglect

A high-risk population was successfully identified by the use of perinatal screening procedures. These children encountered significantly different parenting practices than the low-risk "control" group. There were also significant differences between "High-Risk" and "Low-Risk" families on issues concerning social adjustment and mother/infant attachment. Five children in a "High-Risk Nonintervene" group required hospitalization for serious injuries thought to be secondary to abnormal parenting practices. There were no such hospitalizations in either a High Risk Intervene group or a Low Risk group. Labordelivery observations, and postpartum interviews and observations provided the most accurate predictive information; prenatal interviews and questionnaires did not add significantly. Perinatal assessment and simple intervention with families at high risk for abnormal parenting practices significantly improves the infants' chances for escaping serious hospitalized physical injury.     

Estimators for the Finite Mixture of Rayleigh Model Based on Progressively Censored Data

In this article, based on progressively Type-II censored samples from a heterogeneous population that can be represented by a finite mixture of two-component Rayleigh lifetime model, the problem of estimating the parameters and some lifetime parameters (reliability and hazard functions) are considered. Both Bayesian and maximum likelihood estimators are of interest. A class of natural conjugate prior densities is considered in the Bayesian setting. The Bayes estimators are obtained using both the symmetric (squared error) loss function, and the asymmetric (LINEX and General Entropy) loss functions. It has been seen that the estimators obtained can be easily evaluated for this type of censoring by using suitable numerical methods. Finally, the performance of the estimates have been compared on the basis of their simulated maximum square error via a Monte Carlo simulation study.     

Approximate MLEs of the parameters of location-scale models under type II censoring

Log-location-scale distributions are widely used parametric models that have fundamental importance in both parametric and semiparametric frameworks. The likelihood equations based on a Type II censored sample from location-scale distributions do not provide explicit solutions for the para-meters. Statistical software is widely available and is based on iterative methods (such as, Newton Raphson Algorithm, EM algorithm etc.), which require starting values near the global maximum. There are also many situations that the specialized software does not handle. This paper provides a method for determining explicit estimators for the location and scale parameters by approximating the likelihood function, where the method does not require any starting values. The performance of the proposed approximate method for the Weibull distribution and Log-Logistic distributions is compared with those based on iterative methods through the use of simulation studies for a wide range of sample size and Type II censoring schemes. Here we also examine the probability coverages of the pivotal quantities based on asymptotic normality. In addition, two examples are given.     

Reliability estimation in stress–strength models: an MCMC approach

In this paper, we consider the estimation of the stress–strength parameter R=P(Y<X) when X and Y are independent and both are modified Weibull distributions with the common two shape parameters but different scale parameters. The Markov Chain Monte Carlo sampling method is used for posterior inference of the reliability of the stress–strength model. The maximum-likelihood estimator of R and its asymptotic distribution are obtained. Based on the asymptotic distribution, the confidence interval of R can be obtained using the delta method. We also propose a bootstrap confidence interval of R. The Bayesian estimators with balanced loss function, using informative and non-informative priors, are derived. Different methods and the corresponding confidence intervals are compared using Monte Carlo simulations.     

Predictive Inference and Scientific Reproducibility

ABSTRACT

Most statistical analyses use hypothesis tests or estimation about parameters to form inferential conclusions. I think this is noble, but misguided. The point of view expressed here is that observables are fundamental, and that the goal of statistical modeling should be to predict future observations, given the current data and other relevant information. Further, the prediction of future observables provides multiple advantages to practicing scientists, and to science in general. These include an interpretable numerical summary of a quantity of direct interest to current and future researchers, a calibrated prediction of what’s likely to happen in future experiments, a prediction that can be either “corroborated” or “refuted” through experimentation, and avoidance of inference about parameters; quantities that exists only as convenient indices of hypothetical distributions. Finally, the predictive probability of a future observable can be used as a standard for communicating the reliability of the current work, regardless of whether confirmatory experiments are conducted. Adoption of this paradigm would improve our rigor for scientific accuracy and reproducibility by shifting our focus from “finding differences” among hypothetical parameters to predicting observable events based on our current scientific understanding.     

THERAPIST PHYSICAL ATTRACTIVENESS: AN UNEXPLORED INFLUENCE ON CLIENT DISCLOSURE

Existing research indicates that clients perceive facially attractive therapists as more competent, trustworthy, genuine, and effective than less attractive therapists. No Studies exist to help explain how the therapist's attractiveness influences a client's self-disclosure. Participants (n = 241) were randomly assigned to one of eight experimental groups to test the interaction of the therapist's attractiveness, client's gender, the nature of presenting problem, and the client's comfort with disclosing in a hypothetical couple therapy scenario. Analysis of variance procedures established that most participants reported feeling more comfortable disclosing a benign (communications) problem than a potentially embarrassing (sexual) problem, and more comfortable disclosing problem to an attractive than to a less attractive female therapist. Therapists are encouraged to understand the power attractiveness may have in their own and their client's lives