Estimation and regularization techniques for regression models with multidimensional prediction functions

Boosting is one of the most important methods for fitting regression models and building prediction rules. A notable feature of boosting is that the technique can be modified such that it includes a built-in mechanism for shrinking coefficient estimates and variable selection. This regularization mechanism makes boosting a suitable method for analyzing data characterized by small sample sizes and large numbers of predictors. We extend the existing methodology by developing a boosting method for prediction functions with multiple components. Such multidimensional functions occur in many types of statistical models, for example in count data models and in models involving outcome variables with a mixture distribution. As will be demonstrated, the new algorithm is suitable for both the estimation of the prediction function and regularization of the estimates. In addition, nuisance parameters can be estimated simultaneously with the prediction function.     

The allocation and value of time assigned to housework and child-care: An analysis for Switzerland

In this paper, data from the 1997 Swiss Labour Force Survey are used to analyse the allocation and value of time assigned to housework and child-care. It is shown that men's allocation of time to housework and child-care is largely invariant to changes in socio-economic factors. Women's allocation of time to housework and child-care, on the other hand, is shown to depend on several social, economic, and demographic factors. The value of time assigned to housework and child-care is calculated with two market replacement cost methods and three opportunity cost methods. The results show that the value of time assigned to housework and child-care ranges from 27% to 39% and from 5% to 8% of GDP (in 1997), respectively. The value of time assigned to housework and child-care is also calculated for different household structures. Received: 15 April 1999/Accepted: 5 May 2000     

Estimation of a linear model under microaggregation by individual ranking

Summary Microaggregation by individual ranking is one of themost commonly applied disclosure control techniques for continuous microdata. The paper studies the effect of microaggregation by individual ranking on the least squares estimation of a multiple linear regression model. It is shown that the traditional least squares estimates are asymptotically unbiased. Moreover, the least squares estimates asymptotically have the same variances as the least squares estimates based on the original (non-aggregated) data. Thus, asymptotically, microaggregation by individual ranking does not result in a loss of efficiency in the least squares estimation of a multiple linear regression model. I thank Hans Schneeweiss for very helpful discussions and comments. Financial support from the Deutsche Forschungsgemeinschaft (German Science Foundation) is gratefully acknowledged.     

Computation of maximum likelihood estimates for μ and β from a grouped sample of a normal population. A comparison of algorithms

Computation of Maximum Likelihood Estimates for μ and β from a grouped sample of a normal population is a numerical problem which occurs frequently in applied statistics. We investigate the performance of several algorithms for this problem (including Newton-Raphson, Method of Scoring, Expectation-Maximization and others) using simulated data. One of the main results is that Method of Scoring is best both in number of iterations and CPU time though no second-order derivatives are used.     

Public attitudes toward child maltreatment in Israel: Implications for policy

Objective

The public view of the severity of social problems and their perceptions of how effectively they are being addressed have a major impact on public policies and resource allocation. The present study focuses on public attitudes toward child maltreatment. It examines perceptions of child maltreatment as a social problem, and attitudes toward prevention and treatment strategies in cases of child abuse and neglect.

Methods

A survey was conducted among a representative sample of 812 Israeli adults: 688 Jews, and 124 Arabs. Fifty additional Ultra-Orthodox Jews were added for comparisons among Jewish participants.

Results

The participants tend to view violence and alcohol consumption among youth as a more serious problem than parental maltreatment of children. Low-income participants tend to view child maltreatment as a more serious problem than did high-income participants. Jewish participants (except for ultra-Orthodox Jews) view child maltreatment as a significantly more serious problem than did Arabs and Ultra-Orthodox Jews. Most of the participants believe that parents who maltreat their children should be punished, but they should be taught how to refrain from maltreatment, and learn to change their behavior. Many participants believe that the courts do not sufficiently punish parents who maltreat their children. In their assessments of the quality of professional work, the highest evaluations are given to services provided by physicians and social workers, whereas the lowest evaluations are given to judges and police. Most of the participants (70%) believed that social workers play an important role in protecting children.

Conclusions

The survey findings have implications for enhancing public awareness of child maltreatment as well for policy making in the area of child abuse and neglect. The differences in the perceptions of participants based on their religious, cultural and economic background toward child maltreatment should be further studied and addressed in policy and interventions in this area.     

Teaching anti-oppressive social work

This article highlights critical elements for effective and culturally sensitive teaching when transferring educational programs from a Western country to another context. The Social Work Education Project (SWEP), a master's level professional education program, was developed to train social workers in the conflict-affected area of Mindanao in the Philippines. Designed to enhance the leadership capacity of local social workers and contribute to the peace process, SWEP represents collaboration between the National Catholic School of Social Service in Washington, DC; Community and Family Services International, a Philippines-based NGO; and other Philippine community and educational partners. Qualitative program evaluation findings reveal that students assessed their learning as culturally relevant and personally meaningful while also enhancing their capacities for effective social work practice. Grounded in these findings and the experiences of involved faculty, the article concludes with five key lessons for international education.     

Outlier Robust Small‐Area Estimation Under Spatial Correlation

Modern systems of official statistics require the estimation and publication of business statistics for disaggregated domains, for example, industry domains and geographical regions. Outlier robust methods have proven to be useful for small‐area estimation. Recently proposed outlier robust model‐based small‐area methods assume, however, uncorrelated random effects. Spatial dependencies, resulting from similar industry domains or geographic regions, often occur. In this paper, we propose an outlier robust small‐area methodology that allows for the presence of spatial correlation in the data. In particular, we present a robust predictive methodology that incorporates the potential spatial impact from other areas (domains) on the small area (domain) of interest. We further propose two parametric bootstrap methods for estimating the mean‐squared error. Simulations indicate that the proposed methodology may lead to efficiency gains. The paper concludes with an illustrative application by using business data for estimating average labour costs in Italian provinces.