Master of Science (MS)
First Committee Member
Number of Pages
The problem of estimating exposure to a pollutant involves estimation of chemical intake Q. Typically, Q has the form Q = q11q12&cdots;q 1r1/q211q22 &cdots;q2r2t where all thetaij's are unknown means of certain random variables. In assessing the risk to human health from exposure to the pollutant of concern, a confidence interval estimate of Q is needed. In the case of independent normal random variables, point estimates and certain types of interval estimates of a product of several parameters exist in the literature, but not for a ratio of products. In many situations, lot of prior knowledge is available regarding the random variables corresponding to these parameters. In the case of independent normal random variables, we look at this problem from a Bayesian approach, and show how to incorporate prior knowledge to calculate confidence interval for Q. In situations with no prior knowledge, generalized Bayes approach with noninformative priors can be used. Several real and simulated examples are provided to demonstrate the proposed procedures.
Assessment; Confidence; Exposure; Intervals; Risk
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.
Kirkham, Mark S, "Confidence intervals for exposure risk assessment" (2002). UNLV Retrospective Theses & Dissertations. 1431.