Adaptive SamplingOffering a viable solution to the long-standing problem of estimating the abundance of rare, clustered populations, adaptive sampling designs are rapidly gaining prominence in the natural and social sciences as well as in other fields with inherently difficult sampling situations. In marked contrast to conventional sampling designs, in which the entire sample of units to be observed is fixed prior to the survey, adaptive sampling strategies allow for increased sampling intensity depending upon observations made during the survey. For example, in a survey to assess the abundance of a rare animal species, neighboring sites may be added to the sample whenever the species is encountered during the survey. In an epidemiological survey of a contagious or genetically linked disease, sampling intensity may be increased whenever prevalence of the disease is encountered. Written by two acknowledged experts in this emerging field, this book offers researchers their first comprehensive introduction to adaptive sampling. An ideal reference for statisticians conducting research in survey designs and spatial statistics as well as researchers working in the environmental, ecological, public health, and biomedical sciences. Adaptive Sampling:
|
What people are saying - Write a review
We haven't found any reviews in the usual places.
Contents
Efficiency and Sample Size Issues | 149 |
Adaptive Cluster Sampling Based on Order Statistics | 163 |
Multivariate Aspects of Adaptive Sampling | 201 |
Copyright | |
4 other sections not shown
Common terms and phrases
adaptive cluster sampling adaptive design adaptive sampling added additional allocation Analysis animals apply approach associated assumed Chapter combination components computed conditional consider consists contains conventional design cost defined denote density depend described detectability distinct distribution efficiency equal error example expected Figure final fixed function given gives Hence included independent inference initial sample intersection known labels likelihood linear mean-square method minimal namely neighborhood normal objects observed obtained optimal outcome parameter partition phase plots population total possible predictive predictor primary units probability procedure reduced relative replacement sample mean sample size sampling design satisfying selected simple random sampling situation statistic strata strategy stratified stratum Suppose survey Table term Theorem theory Thompson tree unbiased estimator units values variable variable of interest variance vector y-values