1,720,987 research outputs found
Robust case-weighting for multipurpose establishment surveys.
No full textCase-weighting or assigning a unique weight to each sample unit is a popular method of sample weighting when internal consistency of the survey estimates is paramount. If in addition external constraints on key variables (the survey benchmarks) must also be met, then case-weights computed via generalised least squares, based on an assumed linear regression model for the survey variables, can be used. Unfortunately, this method of weighting can lead to negative case-weights. It is also susceptible to bias if the linear model is misspecified. This article proposes a modified method of linear regression-based case-weighting which ensures positive weights via use of a ridging procedure, and model misspecification robustness via the inclusion of a nonparametric regression bias correction factor. Empirical results which illustrate the gains from the new method of weighting are presente
Analysis of survey data
No full textThis book is concerned with statistical methods for the analysis of data collected from a survey. A survey could consist of data collected from a questionnaire or from measurements, such as those taken as part of a quality control process. Concerned with the statistical methods for the analysis of sample survey data, this book will update and extend the successful book edited by Skinner, Holt and Smith on 'Analysis of Complex Surveys'. The focus will be on methodological issues, which arise when applying statistical methods to sample survey data and will discuss in detail the impact of complex sampling schemes. Further issues, such as how to deal with missing data and measurement of error will also be critically discussed. There have significant improvements in statistical software which implement complex sampling schemes (eg SUDAAN, STATA, WESVAR, PC CARP ) in the last decade and there is greater need for practical advice for those analysing survey data. To ensure a broad audience, the statistical theory will be made accessible through the use of practical examples. This book will be accessible to a broad audience of statisticians but will primarily be of interest to practitioners analysing survey data. Increased awareness by social scientists of the variety of powerful statistical methods will make this book a useful reference
Limited information likelihood analysis of survey data
No full textAnalysts of survey data are often interested in modelling the population process, or superpopulation, that gave rise to a 'target' set of survey variables. An important tool for this is maximum likelihood estimation. A survey is said to provide limited information for such inference if data used in the design of the survey are unavailable to the analyst. In this circumstance, sample inclusion probabilities, which are typically available, provide information which needs to be incorporated into the analysis. We consider the case where these inclusion probabilities can be modelled in terms of a linear combination of the design and target variables, and only sample values of these are available. Strict maximum likelihood estimation of the underlying superpopulation means of these variables appears to be analytically impossible in this case, but an analysis based on approximations to the inclusion probabilities leads to a simple estimator which is a close approximation to the maximum likelihood estimator. In a simulation study, this estimator outperformed several other estimators that are based on approaches suggested in the sampling literature
Methodology for estimating the abundance of rare animals: Seabird nesting on North East Herald Cay
No full textWe discuss the problem of estimating the number of nests of different species of seabirds on North East Herald Cay based on the data from a 1996 survey of quadrats along transects and data from similar past surveys. We consider three approaches based on different plausible models, namely a conditional negative binomial model that allows for additional zeroes in the data, a weighting approach (based on a heteroscedastic regression model), and a transform-both-sides regression approach. We find that the conditional negative binomial approach and a linear regression approach work well but that the transform-both-sides approach should not be used. We apply the conditional negative binomial and linear regression approaches with poststratification based on data quality and availability to estimate the number of frigatebird nests on North East Herald Cay
Outlier robust small area estimation
No full textRecently proposed outlier robust small area estimators can be substantially biased when outliers are drawn from a distribution that has a different mean from that of the rest of the survey data. This naturally leads one to consider an outlier robust bias correction for these estimators. We develop this idea, proposing two different analytical mean-squared error estimators for the ensuing bias-corrected outlier robust estimators. Simulations based on realistic outlier-contaminated data show that the bias correction proposed often leads to more efficient estimators. Furthermore, the mean-squared error estimation methods proposed appear to perform well with a variety of outlier robust small area estimators
- …
