Standard errors for EM estimation

The EM algorithm is a popular method for computing maximum likelihood estimates. One of its drawbacks is that it does not produce standard errors as a by-product. We consider obtaining standard errors by numerical differentiation. Two approaches are considered. The first differentiates the Fisher score vector to yield the Hessian of the log-likelihood. The second differentiates the EM operator and uses an identity that relates its derivative to the Hessian of the log-likelihood. The well-known SEM algorithm uses the second approach. We consider three additional algorithms: one that uses the first approach and two that use the second. We evaluate the complexity and precision of these three and the SEM in algorithm seven examples. The first is a single-parameter example used to give insight. The others are three examples in each of two areas of EM application: Poisson mixture models and the estimation of covariance from incomplete data. The examples show that there are algorithms that are much simpler and more accurate than the SEM algorithm. Hopefully their simplicity will increase the availability of standard error estimates in EM applications. It is shown that, as previously conjectured, a symmetry diagnostic can accurately estimate errors arising from numerical differentiation. Some issues related to the speed of the EM algorithm and algorithms that differentiate the EM operator are identified.     

The Construct of Mindfulness

After mindfulness is defined, a brief history of the research on the topic to date is reviewed. This work essentially falls into three categories: health, business, and education. Considerations of mindlessness as a social issue are then addressed. A brief introduction to the articles in this issue follows. These articles speak to mindfulness as it relates to potential solutions to social problems.     

Mindfulness Research and the Future

We consider some of the ways in which mindfulness theory can provide new solutions to current problems and address potential future problems that may result from changes in demographics and technology. Specific research questions are suggested to see how increasing mindfulness and decreasing mindlessnesscan work to diminish these social problems as they occur specifically in the classroom, the workplace, and the social predicament of getting old. We discuss (a) recasting some of the problems of the elderly into problems of perspective about the elderly, (b) mindful ways of dealing with increased ethnic diversity, (c) differentiating between changing the workplace work and changing the character of the work itself and (d) the advantages and ways of increasingmindfulness in the classroom.     

Suicide and mental health problems among Swedish youth in the wake of the 1990s recession

The purpose of this explorative study is to bring together a variety of Swedish data sources with indicators of youth mental health and living conditions in order to illuminate trends during the last decades, elucidate possible determinants of mental ill health and develop hypotheses to explain the observed trend patterns. The analyses in the study reveal some surprising inconsistencies with respect to the mental health trends among young people during the 1990s. Most striking is the pattern of increasing youth unemployment coinciding with almost inverse trends in the rates of suicide. The secular trends in fatal suicides during this period do not show any increase but some actual decrease in sub‐populations despite high sustained levels of unemployment. In contrast, survey data indicate that the general mental health of youth during this period appears to have deteriorated.     

Testing for No Effect by Cosine Series Methods

The properties of three lack-of-fit tests that are related to non-parametric cosine regression analysis are examined in the context of testing for a constant mean function. Analytic power comparisons of these tests vs a most powerful test are made using intermediate asymptotic relative efficiency. In particular, a data-driven test is produced which is asymptotically as efficient as the most powerful test over a class of alternatives. A small scale simulation experiment is conducted to ascertain the extent that the large sample comparisons are applicable to finite samples.     

Analysis of Competing Risks by Using Bayesian Smoothing

We consider the competing risks set-up. In many practical situations, the conditional probability of the cause of failure given the failure time is of direct interest. We propose to model the competing risks by the overall hazard rate and the conditional probabilities rather than the cause-specific hazards. We adopt a Bayesian smoothing approach for both quantities of interest. Illustrations are given at the end.     

Living Arrangements and Quality of Life Among Chinese Canadian Elders

This paper examines the role of living arrangements in thequality of life of community-dwelling Chinese elders (aged 65 andover) currently residing in Vancouver and Victoria, BritishColumbia. Data are based on a random sample of 830 persons[response rate = 71.5%], who were interviewed in their homes inthe language of their choice in 1995–96. Three dimensions ofquality of life – satisfaction, well-being and social support –are examined for married men and women [living with spouse alonevs. living intergenerationally] and widowed women [living alonevs. living intergenerationally]. Few differences are found formarried persons, especially women; for widows, living alonesignificantly reduces quality of life in a number of areas.Regression analyses indicate that living arrangements are not asignificant predictor of life satisfaction or well-being formarried men and women. For widows, living arrangements determinewell-being but not life satisfaction.Overall, age, health status, and social support (havingfriends/confidante) are better predictors of quality of life forelderly Chinese Canadians than are living arrangements. Findingshighlight the importance of: empirically distinguishing maritalstatus and living arrangements in studying the quality of life ofelders; not homogenizing Chinese Canadian seniors with regard toliving arrangements; and focussing on Chinese elderly widows wholive alone as a group at risk of low well-being.     

The success of protest groups: Multivariate analyses

This paper reviews earlier research and presents new analytical findings regarding the outcomes of social movements. Using the resource mobilization/management approach, empirical propositions that seek to explain protest group success or failure are tested. Based upon data gathered from a sample of 53 US protest groups, the causal models explained the majority of the variance in degree of success between these groups. Our findings indicate that protest groups which threaten to replace or destroy established groups are usually unsuccessful, and those having many strong alliances tend to be more successful than groups fighting alone. The use of violence does not greatly aid the prediction of group outcome because of the unpredictable, ambivalent reaction to violence by established groups.     

Care of the Elderly—A Project,a Protest,a Survey and a Plan

Rehousing of elderly people in Australia has generally not been preceded by research into the kind of assistance that is necessary. An example is given of how an attempt was made to discover the needs of the elderly disabled in a small country community. Assessment by observation of people at home revealed the need for assistance for them to stay where they were. Home Care organized from the district hospital is often a more appropriate answer to this problem than the building of expensive and unnecessary institutions.