Application of adaptive sampling in fishery part 2: truncated adaptive cluster sampling designs
There are some experiences that researcher come across quite number of time for very large networks in the initial samples such that they cannot finish the sampling procedure. Two solutions have been proposed and used by marine biologists which we discuss in this article: i) Adaptive cluster sampling based on order statistics with a stopping rule, ii) Restricted adaptive cluster sampling. Until recently, the unbiased estimators were not available for both sampling. Restricted adaptive cluster sampling was used (Lo et al., 1997) under investigation in US Southwest Fisheries Science Center which was reasonably efficient even with a biased estimator. Salehi and Seber (2002) propose an unbiased estimator for this sampling design. They show that the unbiased estimator is shown to compare very favorably with the standard biased estimators, using simulation.
| Main Author: | |
|---|---|
| Format: | article biblioteca |
| Language: | English |
| Published: |
2001
|
| Subjects: | Biology, Fisheries, Sampling design, Clump fish population, Abundance estimation, Iran, |
| Online Access: | http://hdl.handle.net/1834/37034 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | There are some experiences that researcher come across quite number of time for very large networks in the initial samples such that they cannot finish the sampling procedure. Two solutions have been proposed and used by marine biologists which we discuss in this article: i) Adaptive cluster sampling based on order statistics with a stopping rule, ii) Restricted adaptive cluster sampling. Until recently, the unbiased estimators were not available for both sampling. Restricted adaptive cluster sampling was used (Lo et al., 1997) under investigation in US Southwest Fisheries Science Center which was reasonably efficient even with a biased estimator. Salehi and Seber (2002) propose an unbiased estimator for this sampling design. They show that the unbiased estimator is shown to compare very favorably with the standard biased estimators, using simulation. |
|---|