this paper I discuss a Bayesian approach to solving this problem that has long been available in principle but is only now becoming routinely feasible, by virtue of recent computational advances, and examine its implementation in examples that involve forecasting the price of oil and estimating the chance of catastrophic failure of the U.S. Space Shuttle.

Abstract not found

We present a method of analyzing a series of independent cross-sectional surveys in which some questions are not answered in some surveys and some respondents do not answer some of the questions posed. The method is also applicable to a single survey in which different questions are asked, or different sampling methods used, in different strata or clusters. Our method involves multiply-imputing the missing items and questions by adding to existing methods of imputation designed for single surveys a hierarchical regression model that allows covariates at the individual and survey levels. Information from survey weights is exploited by including in the analysis the variables on which the weights were based, and then reweighting individual responses (observed and imputed) to estimate population quantities. We also develop diagnostics for checking the fit of the imputation model based on comparing imputed to nonimputed data. We illustrate with the example that motivated this project --- a ...

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

This paper addresses the use of weight functions to model publication bias in meta-analysis. Since this bias is hard to gauge, we introduce a non-parametric "-contamination class of weight functions. We then illustrate how to explore sensitivity of conclusions to the specification of the weight function by examining the range of results for the entire class. We find lower bounds on the coverage of confidence intervals. If no publication bias is present, results are robust even when considered over the entire "-contamination class. However, if publication bias is present, then the coverage provided by the usual interval estimator is not robust. In this case, an alternative interval estimator is suggested. We also illustrate how both upper and lower bounds on posterior quantities of interest may be found for the case in which prior information is available. Some key words: Weight functions; Selection bias; Meta-analysis. 1 Introduction This paper addresses the use of weight functions t...

A standard method for reducing survey nonresponse is to offer incentives to survey participants. We reanalyze the data from a recent meta-analysis (Singer et al., 1999), using a hierarchical regression model, in order to estimate the effects of prepaid and postpaid incentives for face-to-face and telephone surveys. We then apply the results of our fitted model to the New York City Social Indicators Survey, a biennial telephone survey with a high nonresponse rate. We consider the balance of estimated costs, cost savings, and response rate for different choices of incentives. In general, we propose a methodology for evaluating some cost-benefit decisions in sample survey design. Keywords: decision analysis, hierarchical linear regression, informal Bayes, meta-analysis, survey nonresponse 1 Introduction What with over 90% of U.S. households having telephones, and the convenience of random digit dialing and computer assisted interviewing, telephone interviewing has become increa...

In wanting to summarise evidence from a number of studies a variety of statistical methods have been proposed. Of these the most widely used is the so-called fixed effect model in which the individual studies are estimating a single, but unknown, overall population effect. When there is `considerable' heterogeneity, in terms of the effect sizes, between the studies the use of a random effect model has been advocated in which each individual study is assumed to be estimating its own, unknown, true effect. Discrimination between fixed and random effect models has been advocated by means of a Ø 2 test for heterogeneity, which it is accepted has low statistical power. Recent interest has been shown in the use of Bayes Factors as an alternative. The use of Bayes factors is illustrated using a number of previously published meta-analyses in which there are varying degrees of heterogeneity. It is shown how the use of Bayes Factors leads to a more intuitive assessment of the evidence in favo...

The hierarchical Bayesian method combines and summarizes the results from multiple sites using a weighted regression analysis where the unit of observation is the site, and the covariates are characteristics of the sites. There are two sources of random error in this model: the usual within-site sampling error and an additional random effect due to unpredictable differences among sites. The hierarchical model allows sites with small sample sizes to "borrow strength" from the others, to the extent that the between-site variance is estimated to be small. A special graph, called a trace plot, displays the posterior distribution of the among-site standard deviation and its effect on the other parameter estimates. This paper combines and re-analyzes data on schizophrenic births collected across seventeen states of the U.S. We found additional evidence that severe weather may be associated with increased risk of schizophrenia and confirmed the existence of a dose-response curve, where the risk for schizophrenia increases as severity of winter increases.

this article we attempt to provide an approach to quantifying the costs and benefits of incentives, as a means of assisting in the decision of whether and how to apply an incentive in a particular telephone survey. We proceed in two steps. In Section 2 we reanalyze the data from the comprehensive metaanalysis of Singer, Van Hoewyk, Gebler, Raghunathan, and McGonagle (1999) of incentives in face-to-face and telephone surveys and model the effect of incentives on response rates as a function of timing and amount of incentive and descriptors of the survey. In Section 3 we apply the model estimates to the cost structure of the New York City Social Indicators Survey, a biennial study with a nonresponse rate in the 50% range. In Section 4 we consider how these ideas can be applied generally and discuss limitations of our approach

In this paper we develop a hierarchical bivariate time series model to characterize the relationship between particulate matter less than 10 microns in aerodynamic diameter (P M10) and both mortality and hospital admissions for cardiovascular diseases. The model is applied to time series data on mortality and morbidity for 10 metropolitan areas in the United States from 1986 to 1993. We postulate that these time series should be related through a shared relationship with P M10. At the first stage of the hierarchy, we fit two seemingly unrelated Poisson regression models to produce city-specific estimates of the log relative rates of mortality and morbidity associated with exposure to P M10 within each location. The sample covariance matrix of the estimated log relative rates is obtained using a novel generalized estimating equation approach that takes into account the correlation between the mortality and morbidity time series. At the second stage, we combine information across locations to estimate overall log relative rates of mortality and morbidity and variation of the rates across cities. Using the combined information across the 10 locations we find that a 10 µg/m3 increase in average P M10 at the current day and previous day is associated with a 0.26 % increase in mortality