An investigation of the impact of left‐censored soil contamination data on the uncertainty of descriptive statistical parameters
Left‐censored concentration data are frequently encountered because measuring instruments cannot detect concentrations below instruments detection limit (DL). For statistical analysis of left‐censored data, environmental literature mainly refers to the following methods: maximum likelihood estimation (MLE), regression on order statistics using lognormal and gamma assumption (rROS and GROS, respectively), and Kaplan‐Meier. A number of simulation studies examined the performance of these methods in terms of bias and/or mean square error. However, no matter which method is adopted, some uncertainty is introduced into outcomes since all is known about a left‐censored observation is that the concentration falls between 0 and the DL. Data used here come from analysis of soil samples collected for a site characterization in Montreal, Canada. Employing nonparametric bootstrap, the authors quantify the uncertainty and bias in the mean and standard deviation estimates obtained by the MLE (under lognormal, Weibull, and gamma distributions), rROS, GROS, and KM methods. First, the authors demonstrate that the highest uncertainty is associated with MLEs under lognormality and Weibull assumptions while a gamma assumption leads to estimates with less uncertainty. Second, the authors show that although an increase in sample size improves the uncertainty, it reduces the bias only in the rROS, GROS, and KM methods. Finally, comparing percentage uncertainty in the mean of contaminant data, the authors illustrate that adopting an inappropriate estimator results in large uncertainties. This article is protected by copyright. All rights reserved
Customer comments
No comments were found for An investigation of the impact of left‐censored soil contamination data on the uncertainty of descriptive statistical parameters. Be the first to comment!