Validation of a longitudinally measured surrogate marker for a time-to-event endpoint

The objective of this paper is to extend the surrogate endpoint validation methodology proposed by Buyse et al. (2000) to the case of a longitudinally measured surrogate marker when the endpoint of interest is time to some key clinical event. A joint model for longitudinal and event time data is required. To this end, the model formulation of Henderson et al. (2000) is adopted. The methodology is applied to a set of two randomized clinical trials in advanced prostate cancer to evaluate the usefulness of prostate-specific antigen (PSA) level as a surrogate for survival.     

The structure and significance of community satisfaction

Residential and community satisfaction have been studied for many decades in an effort to understand human commitments to local places. In more recent years, concepts that more nearly reflect commitment have been developed, among them place attachment. Yet, the extensive, prior work on empirical indicators of satisfaction give that concept a particularly useful entry into studies of experiences of place, even though these may only provide estimates of phenomena of deeper commitment. The present study is drawn from the data for a large-scale investigation of the meaning ofcommunity in metropolitan America. Based on 2,622 women and men from 42 municipalities in 10 SMSAs across the United States, the study includes a sample of all socioeconomic statuses and from minority as well as majority populations. On the basis of factor analyses of a wide array of evaluations and satisfactions with housing, neighborhood, and the wider community, a structure for community satisfaction has been delineated. On the basis of these community satisfaction factors, along with several other social role and personality variables, an estimate is provided of the personal significance of community satisfaction in people's lives, using life satisfaction as the dependent variable. While there are socioeconomic status and other variations, community satisfactions appear to be important influences on life satisfaction at all socioeconomic status levels.Requests for reprints should be addressed to Marc Fried, Dept. of Psychology, Boston College, Chestnut Hill, MA 02167.The now defunct Center for Metropolitan Studies of the National Institute of Mental Health was most helpful, not only in funding this study but in many other respects as well. The author is also grateful to the many people who participated in the large-scale study of which this paper represents only a small part. In particular, John Havens and Lorna Ferguson were most important to the achieving of the objectives of the study.     

RANDOM SAMPLING IN PATTERN RECONSTRUCTION

A random pattern is the realization of a random set process giving a partition of a set. We are interested by the situation where an unknown random pattern has to be estimated (reconstructed) from sample data. Two sampling strategies to do so are presented. One is of random character the other is not. The risk corresponding to each is obtained. Some numerical comparisons are made for a particular model.     

Homelessness in Europe and the United States: A Comparison of Prevalence and Public Opinion

Random samples of 250–435 adults were interviewed by telephone in five different nations ( N = 1,546): Belgium, Germany, Italy, the UK, and the United States. The interview included questions on respondent attitudes, knowledge, and opinions regarding homelessness; respondents' own personal experiences with homelessness and homeless people; and demographic characteristics of the respondents. The highest rates for lifetime literal homelessness were found in the UK (7.7%) and United States (6.2%), with the lowest rate in Germany (2.4%), and intermediate rates in Italy (4.0%) and Belgium (3.4%). Less compassionate attitudes toward the homeless were also found on many dimensions in the United States and the UK. Possible explanations of these findings, drawn from various theoretical perspectives, and policy implications are provided .     

A Dynamic Model of Welfare Reform

A dynamic structural model of labor supply, welfare participation, and food stamp participation is estimated using the 1992, 1993, and 1996 panels of the Survey of Income and Program Participation. Details of various policies including welfare time limits, work requirements, and Earned Income Tax Credit (EITC) are incorporated formally in the budget constraint. Policy simulations reveal that the economy accounts for half of the increase in the labor supply of female heads of family between 1992 and 1999. A time limit results in a larger efficiency gain than a work requirement or a direct reduction in welfare benefits. A reform package can lead to both a reduction in the government expenditure and an improvement in utility. The EITC expansion results in a substantial efficiency gain among individuals with the lowest expected wage. These individuals are almost unaffected by the economic expansion, but their income and utility increase significantly under the reform package.     

VOLATILITY,LABOR MARKET FLEXIBILITY,AND THE PATTERN OF COMPARATIVE ADVANTAGE

This paper studies the link between volatility, labor market flexibility, and international trade. International differences in labor market regulations affect how firms can adjust to idiosyncratic shocks. These institutional differences interact with sector specific differences in volatility (the variance of the firm‐specific shocks in a sector) to generate a new source of comparative advantage. Other things equal, countries with more flexible labor markets specialize in sectors with higher volatility. Empirical evidence for a large sample of countries strongly supports this theory: the exports of countries with more flexible labor markets are biased towards high‐volatility sectors. We show how differences in labor market institutions can be parsimoniously integrated into the workhorse model of Ricardian comparative advantage of Dornbusch, Fischer, and Samuelson (1977, American Economic Review, 67, 823–839). We also show how our model can be extended to multiple factors of production.