Bootstrap techniques for proportional hazards models with censored observations

We propose a bootstrap technique for generating pseudo-samples from survival data containing censored observations. This simulation selects a survival time with replacement from the data and then assigns a covariate according to the model of proportional hazards. We also develop a constrained bootstrap technique in which every pseudo-sample has the same distribution of covariate values as does the original, observed data. We use these simulation techniques to estimate the bias and variance of regression coefficients and to approximate the significance levels of goodness-of-fit statistics for testing the assumption of the proportional hazards model.     

How Children Spend Their Time: A Sample Survey for Use in Exposure and Risk Assessments

Children are becoming an increasingly important focus for exposure and risk assessments because they are more sensitive than adults to environmental contaminants. A necessary step in measuring the extent of children's exposure and in calculating risk assessments is to document how and where children spend their time. This 1990-1991 survey of 1000 households was designed for this purpose, targeting children between 5 and 12 years of age, in six states in varied geographic regions. The behavior of children was sampled on both weekdays and weekends over all four seasons of the year using a retrospective time diary to allocate time to activities during the previous 24 h. Information was obtained on the kinds and locations of activities, the nature of the microenvironments of the locations, and the time spent in the different environments. Measures of variability in addition to mean hours per day are reported. Results of this study closely match those of earlier research on California children's activities done by the California Air Resources Board. One important finding of the survey was that 5- to 12-year-old children in all geographic regions spend most of their time indoors at home, indicating that risk assessments should focus on indoor, on-site hazards.     

What do teenagers want? What do teenagers need?

It is widely accepted that adolescence is a developmental phase fraught with challenges, yet in the 1990s, teenagers face unique demands and difficulties. Dilemmas such as increased violence, drug and alcohol use, sexual involvement and AIDS pose problems not only for the teens themselves, but also for their families, teachers, health care providers and other caregivers.This article reports on the qualitative findings extracted from a study of over 3,700 teenagers from urban, suburban, and small city settings, describing the array of difficulties teenagers face today. The themes which emerged from the students' comments offer a compelling portrait of life as a teenager in the 1990s and help shape an agenda for understanding and responding to these myriad needsThis research was supported by grants from the Ohio Department of Mental Health (#93,1058A), Treu-Mart Fund, Nord Family Foundation, and the University at Albany Faculty Research Awards Program.     

Working with parents of spinal cord injured adolescents: A family systems perspective

Concepts from family systems theory are used to facilitate understanding of the needs of families coping with the spinal cord injury of an adolescent. Five assumptions derived from systemic principles are discussed as related to assessment and family functioning. Related suggestions are given for assisting parents in dealing with this non-normative stressor and its long-range effects on the adolescent and family.     

教育研究中比较的层次问题

本文认为比较教育传统上以国别和世界地区为分析单位容易导致不平衡和不全面的观点，指出有必要建立更加宽泛的比较教育概念范围，提倡并论证了多层次比较分析作为一种研究形式所具备的综合整理各单一层次分析结果的价值所在。     

Revisiting the flexibility and regularity properties of the asymptotically ideal production model

In this paper we estimate the flexibility properties and regular regions for the first three orders of the seminonparametric Asymptotically Ideal Model (AIM) under four different types of constant returns to scale production technologies. The AIM model's parameters are estimatedfrom a Monte Carlo simulation where the data is generated from a three input, one output, globally regular Constant Differences of Elasticity of Substitution function. The Monte Carlo's input quantity and elasticity of substitution estimates at the unit vector are graphed along with the area of the relative price space where the AIM model is monotonically increasing andquasi-concave.     

Stealing from thy Neighbour: Leveraged Recapitalizations and Wealth Redistribution

Leverage-increasing transactions such as recapitalizations and leveraged buy-outs are known to create shareholder wealth. It has been argued that the only sources of this value are efficiency improvements and tax benefits brought about by the higher debt load. This paper shows empirically that contrary to popular belief a portion of this wealth comes at the expense of current bondholders. It is estimated that between 20 per cent and SO per cent of the value created during a leveraged recapitalization can be attributed to wealth transfers from bondholders to shareholders.