Search

A Bayesian approach is presented for estimating a mixture of linear Gaussian stale space models. Such models are used to model interventions in time series and nonparametric regression. Markov chain Monte Carlo sampling is usually necessary to obtain the posterior distributions of such mixture models, because it is difficult to obtain them analytically. The methodological contribution of the article is to derive a set of recursions for dynamic mixture models that efficiently implement a Markov chain Monte Carlo sampling scheme that converges rapidly to the posterior distribution. The methodology is illustrated by fitting an autoregressive model subject to interventions to zinc concentration in sludge.

Shape-memory behavior ls the ability of ccrwin materials to recover, on heating, apparently plastic deformation sustained below a critical temperature. Some materials have good shape-memory behavior as single crystals but little or none as polycrystals, while others have good shape-memory behavior even as polycrystals. Bhattacharya and Kohn (1996. 1997) have proposed a framework to understand this difference. They use energy minimization and the Taylor estimate to argue that the recoverable strains in a polycrystal depend not only on the texture of the polycrystal and the transformation, but critically on the change in symmetry during the underlying martensitic phase transformation. Their results agree with the experimental observations. Shu and Bhattacharya (1997) have also used the Taylor estimate to study the effect of texture in polycrys- tals of Nickel-Titanium and Copper based shape-memory alloys. The use of the Taylor estimate was evaluated in some detail in Bhattacharya and Kohn ( 1997) and more recently in Shu and Bhattacharya (1997) and Shu (1997). In this short report, we summarize the model of recoverable strain and discuss some results that allow us to evaluate the Taylor estimate.

Few empirical studies have examined the attitudes and ethical beliefs of psychiatrists. In this study, 328 members of the Swedish Psychiatric Association were randomly selected to respond to a questionnaire containing three clinical vignettes examining involuntary hospitalization, other compulsory interventions, and restraint. The questionnaire also contained 16 controversial statements and items on the abuse of psychiatry. A 60% response rate was obtained. Psychiatrists' responses were influenced not only by the severity of and risks associated with the patient's disorder, but also by family pressure which affected the decision-making process. Female psychiatrists less often suggested the use of physical restraints and the compulsory use of ECT. Sexual misconduct, and also the inadequate treatment of refugees, were the most frequently reported forms of abuse of psychiatry.

We present a Bayesian approach for estimating nonparametrically an additive autoregressive model with the regression curve estimates cubic smoothing splines. Our approach is robust to innovation outliers; it can handle missing observations and produce multistep ahead forecasts. The computation is carried out using Markov chain Monte Carlo and requires O(nM) operations where n is the sample size and M is the number of Markov chain iterations. This makes it the first exact algorithm for spline smoothing of an additive autoregressive model which can handle large data sets. The properties of the estimates and forecasts are studied empirically using simulated and real data sets.