Regency EF
Monotonic Nonparametric Dose Response Model (303998)
*Faten Alamri , Princess Nourah bint Abdulrahman University/Virginia Commonwealth UniversityKeywords: Tolerable region, Nonparametric modeling, Monotonic basis function, Toxicology data,
There are significant risks in public health caused by environmental exposure to pesticides and many industrial chemicals. This motivates researchers to aim for lower chemical dosage and decreased side effects on our lives. Dose response models explains the relationship between the chemical dosage and the side effects associated with chemical exposure. In particular, agriculture has a high risk of exposure due to fertilizers and pesticides, which are known to contain hazardous chemicals. Toxicologists are interested in determining a proper treatment (dose) of chemicals to eliminate pests that still permits a safe work environment. This work develops a statistical framework to determine the tolerable dose region with low doses of chemicals and minimal side effects using a Nonparametric Bayesian approach. We developed a novel approach to find the tolerable region with a nonparametric curve using monotonic splines by developing a new set of basis functions. The new basis is specifically designed for percent to control data as it ensures that the estimated curve attains the value one at control and is monotonic. The method is illustrated using a dataset from the US EPA.