Some applications of nonlinear and non-Gaussian state–space modelling by means of hidden Markov models

Nonlinear and non-Gaussian state–space models (SSMs) are fitted to different types of time series. The applications include homogeneous and seasonal time series, in particular earthquake counts, polio counts, rainfall occurrence data, glacial varve data and daily returns on a share. The considered SSMs comprise Poisson, Bernoulli, gamma and Student-t distributions at the observation level. Parameter estimations for the SSMs are carried out using a likelihood approximation that is obtained after discretization of the state space. The approximation can be made arbitrarily accurate, and the approximated likelihood is precisely that of a finite-state hidden Markov model (HMM). The proposed method enables us to apply standard HMM techniques. It is easy to implement and can be extended to all kinds of SSMs in a straightforward manner.     

Robust variable selection with application to quality of life research

A large database containing socioeconomic data from 60 communities in Austria and Germany has been built, stemming from 18,000 citizens’ responses to a survey, together with data from official statistical institutes about these communities. This paper describes a procedure for extracting a small set of explanatory variables to explain response variables such as the cognition of quality of life. For better interpretability, the set of explanatory variables needs to be very small and the dependencies among the selected variables need to be low. Due to possible inhomogeneities within the data set, it is further required that the solution is robust to outliers and deviating points. In order to achieve these goals, a robust model selection method, combined with a strategy to reduce the number of selected predictor variables to a necessary minimum, is developed. In addition, this context-sensitive method is applied to obtain responsible factors describing quality of life in communities.     

Robust nonparametric tests for the two-sample location problem

We construct and investigate robust nonparametric tests for the two-sample location problem. A test based on a suitable scaling of the median of the set of differences between the two samples, which is the Hodges-Lehmann shift estimator corresponding to the Wilcoxon two-sample rank test, leads to higher robustness against outliers than the Wilcoxon test itself, while preserving its efficiency under a broad range of distributions. The good performance of the constructed test is investigated under different distributions and outlier configurations and compared to alternatives like the two-sample t-, the Wilcoxon and the median test, as well as to tests based on the difference of the sample medians or the one-sample Hodges-Lehmann estimators.     

Bilanzmanipulation und Managementpräferenzen bezüglich der Strenge des Systems der Rechnungslegungsregulierung

The present paper analyses the political economy of accounting regulation. It develops a game theoretic model of earnings management in a capital market context. Various forms of information asymmetry are examined. The analysis shows that the preferences of managers over the strictness of accounting regulation depend on the nature of the information asymmetry. In a number of instances, managers prefer to be subject to strict regulation. This result contradicts the often heard assertion that managers prefer lax regulation and has implications for regulatory competition.     

格林伯格现代主义绘画的还原论批判

随着美国极简主义艺术的兴起,人们发现形式主义已经走到了最残酷无情的绝对状态,与此同时,也看到了它在艺术上已经走进了一个彻底的死胡同。这种 20世纪 60年代中期激进还原的艺术得出了格林伯格学说的逻辑结论,即现代主义绘画不可避免地走向绝对平面性,而现代主义雕塑则不可避免地走向绝对立体感的清晰度。本人的批评既是对格林伯格现代主义绘画理论的批评,也是对极简主义的格林伯格式鼓吹的批判。     

Comparison of multiple imputation and two-phase logistic regression to analyse two-phase case–control studies with rich phase 1: a simulation study

Two-phase case–control studies cope with the problem of confounding by obtaining required additional information for a subset (phase 2) of all individuals (phase 1). Nowadays, studies with rich phase 1 data are available where only few unmeasured confounders need to be obtained in phase 2. The extended conditional maximum likelihood (ECML) approach in two-phase logistic regression is a novel method to analyse such data. Alternatively, two-phase case–control studies can be analysed by multiple imputation (MI), where phase 2 information for individuals included in phase 1 is treated as missing. We conducted a simulation of two-phase studies, where we compared the performance of ECML and MI in typical scenarios with rich phase 1. Regarding exposure effect, MI was less biased and more precise than ECML. Furthermore, ECML was sensitive against misspecification of the participation model. We therefore recommend MI to analyse two-phase case–control studies in situations with rich phase 1 data.     

Application of Social Science Knowledge to the Community Organization Field

Community organization graduates of the School of Social Work, University of Washington, from 1966 to 1971, found a ready market for their skills although there was evidence of some slackening of demand and broader diversification of initial employment settings in the last two years. Graduates moved rapidly into administrative and management levels and appeared relatively stable in terms of interagency and geographical mobility. The small number of ethnically identified graduates made meaningful comparisons difficult. Sex, however, was a disadvantaging factor in employability and subsequent status. Of the ninety-one graduates surveyed, 94.5 percent responded Similar follow-up study of graduates in other geographic regions was recommended     

State standards for domestic violence perpetrator treatment: current status, trends, and recommendations

We empirically surveyed and analyzed existing standards for the treatment of perpetrators of domestic violence across the United States. Specific areas examined included: presence and scope; administrative entity for certifying; screening and risk assessment protocols; minimum length of treatment; theoretical or conceptual orientation; treatment content; preferred or allowable modalities of treatment; whether research findings are mentioned; methods for revising standards; and minimum education and training required for providers. We examined trends using several methods including comparisons between present and previous survey data (Maiuro et al., 2001). Positive trends were evident including increased use of multivariate models of treatment content, use of an intake assessment prior to treatment, use of a danger/lethality assessment to manage risk, recognition of the need for program evaluation and supportive research, and the requirement of a minimum level of formal education as a prerequisite for providers. We identify specific areas for further research and development and make recommendations for improving existing practice and standards of care.