Bootstrap minimum cost estimation of the average chemical concentration in contaminated soils
✍ Scribed by Ling Chen; Jun Shao
- Publisher
- John Wiley and Sons
- Year
- 1999
- Tongue
- English
- Weight
- 116 KB
- Volume
- 10
- Category
- Article
- ISSN
- 1180-4009
No coin nor oath required. For personal study only.
✦ Synopsis
Contaminated soil is known to have a great impact on the environment and human health. The U.S. Environmental Protection Agency has made eorts to clean up the most hazardous toxic waste sites when the average chemical concentration of the contaminants is higher than a given level. The customary estimate of the average chemical concentration, the sample mean, has a very high chance to underestimate the health risk, since the distribution of chemical concentrations is often positively skewed. We propose a minimum cost estimator which is computed using the bootstrap approximation to the sampling distribution of the t-statistic. The proposed estimator has nice properties (both theoretically and empirically), and is very comparable to (and slightly better than) the minimum cost estimator computed using the Edgeworth expansion. The main advantage of using the bootstrap over the Edgeworth expansion is that the bootstrap does not require complicated theoretical derivations of the Edgeworth expansions so that it can be applied to more complex data such as regression data and/or spatial data.