Master of Science (MS)
First Committee Member
Ashok K. Singh
Number of Pages
Contaminant concentration data from Superfund sites is quite often positively skewed, and the log-normal theory based statistical procedures are typically used for such data. Recent work in the environmental statistics literature, however, has shown that the use of log-normal theory based formulas, such as the H-statistic confidence interval, is problematic. The performance of the H -- UCL in the presence of non -- detects in the sample is investigated via simulated examples. When comparing mean contaminant concentration at a site with that of the background, the 2-sample t-test on log-transformed data is commonly used. A part of this thesis deals with investigation of power of the t-test on log-transformed data by using Monte Carlo simulation.
Analysis; Data; Distribution; Environmental; Lognormal
Statistics; Environmental engineering
University of Nevada, Las Vegas
If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to firstname.lastname@example.org and include clear identification of the work, preferably with URL.
Pant, Devarshi, "On the use of lognormal distribution for environmental data analysis" (2007). UNLV Retrospective Theses & Dissertations. 2204.
IN COPYRIGHT. For more information about this rights statement, please visit http://rightsstatements.org/vocab/InC/1.0/