RISK ANALYSIS IN MORE THAN ONE DIMENSION

The concept of risk function is extended to the situation where the “life” of the specimen is represented by a vector of dimension greater than one. Model-building, estimation of parameters and goodness of fit are discussed. The model is applied to fatigue failure of reinforced concrete beams where two separate indicators of damage are monitored.     

New directions for management accounting

Using a general systems rationale, this paper develops a theoretical structure for approaching the problem of management accounting. The management control problem is explicated in terms of maintaining a relationship between the enterprise's structure and its environment. An enterprise's structure is composed of three elements (and their inter-relations): a decision and control subsystem; a financial funds subsystem and an operating (physical transforms) subsystem. Portrayed in these terms, the problem of management accounting is shown to require a methodology which is able to take cognizance of economic, sociological, psychological and other aspects of the enterprise system. The model described here provides a general intellectual frame of reference for ordering the problem.     

Estimation in Epidemics with Incomplete Observations

The construction of estimating equations by martingale methods is generalized to yield estimators with explicit expressions for the parameters of the birth-and-death and the general epidemic processes when only partial observations are available. (For the birth-and-death process the death process is observed but the number of births is observed only at the end and for the general epidemic process only the removal process is observed.) For large populations, the use of the martingale central limit theorem yields asymptotic confidence regions for the parameters. Explicit expressions are derived for estimators of the variances of the large sample distributions. The range of validity and usefulness of the new estimators is determined by simulation.     

Researching social care for minority ethnic older people: implications of some Scottish research

Summary The paper explores issues raised by research on patterns ofwelfare and mutual care among older Pakistani people and theirfamilies, with particular reference to their use, or lack ofuse of social care services. Qualitative interviews were conductedwith (self-defined) minority ethnic older people, professionalsin statutory services and in minority-led community-based groups.The organization of services is discussed, with special referenceto issues of power and control, the role of the voluntary (non-profit)and statutory sectors, the debate regarding the possible needfor specialist versus generic services, and the role of minority-ledinitiatives. Secondly, issues of service staffing are examined,and the dangers of marginalizing minority services and staffreviewed, especially with reference to divisions of labour inservice provision and staff's own perspectives. Thirdly, themethodological implications of the work are explores, includingissues of linguistic and cultural communication and workingwith relatively powerless social groups, whose views are sought.Examining the views of professional social care workers is alsoproblematized, with reference to their gate-keeping role. Itis argued that user perspectives are essential to the developmentof appropriate services for minority ethnic groups. In conclusion,the need to raise these issues in other work is discussed.     

Estimation for Type II domain of attraction based on the W statistic

The paper presents an estimating equation approach to the estimation of high quantiles of a distribution in the Type II domain of attraction based on the k largest order statistics. The estimators are shown to be consistent. The method fits neatly into a general scheme for estimating high quantiles irrespective of the domain of attraction, which includes Wang's approach to optimally choosing k .     

Hormone profiles,body mass index and aging male symptoms: results of the Androx Vienna Municipality study

Aging in the male is accompanied by steroid hormonal decline, and men may develop symptoms associated with hypogonadism. Increased awareness of ‘andropause’ in recent years has led to greater demand for hormonal assessments, resulting in a rising burden for health economics. We conducted a cross-sectional study to define men at risk for hypogonadism, in whom further hormonal investigation should be performed.

We examined 664 blue-collar workers aged 40–60 years at their workplace and determined hormonal status and body mass index (BMI). Men with an abnormal urogenital status and those on medication that might affect endocrine status were excluded from the study. All participants completed the validated Aging Male Symptom (AMS) questionnaire and obtained scores for psychological symptoms, somatovegetative symptoms, and sexual symptoms.

Multiple logistic regression analyses revealed a significantly increased risk (represented by the odds ratio) of psychological symptoms for men with low levels of testosterone and/or bioavailable testosterone (BAT). Increased BMI as well as low testosterone levels and/or low BAT levels raised the risk of somatovegetative symptoms. Each decrease of BAT by 1?ng/ml caused an approximately 1.8-fold increase of the risk (odds ratio?=?1.832, p?=?0.005). Additional independent risk factors were increased age and low luteinizing hormone (LH) level. Men aged 55 years with BMI >?28?kg/m² and with somatovegetative symptoms and moderate or severe psychological symptoms had a 7.2-fold increase in the risk of a BAT level <?1.5?ng/ml compared to men without these risk factors (p <?0.001). Sensitivity and specificity were 75% and 71%, respectively.

The AMS score combined with age and BMI provides an easy and convenient method to identify men with probable androgen deficiency who require hormonal assessment.     

ESTIMATION OF QUANTILES FOR A FRÉCHET-TYPE DISTRIBUTION

Estimation of high quantiles of a distribution in the domain of attraction of the Fréchet distribution is based on the extremal distribution of the k largest order statistics. The problem is treated by a local maximum likelihood method on a three parameter model. The estimators are shown to be asymptotically consistent for the whole range of the tail index parameter.