Statistical inferences from serially correlated methylene chloride data

While physiologically-based pharmacokinetic (PBPK) modeling has become an important tool in environmental health risk assessment, a rigorous statistical methodology is not routinely incorporated into these analyses. This paper illustrates how to carry out a formal statistical analysis to obtain improved inference upon model parameters from time-dependent, serially correlated methylene chloride data combined from several experiments. Frequentist and Bayesian methods are both considered. We work with a well established PBPK model for disposition of inhaled dichloromethane (DCM, methylene chloride) gas within a closed chamber. 2013, Indian Statistical Institute.

Overview