Dissertation/Thesis Abstract

Pollution Hazard Assessment Using Gamma Regression, Decision Trees, and Hierarchical Modeling
by Lee, Eva, M.S., California State University, Long Beach, 2018, 67; 10973649
Abstract (Summary)

Human health risk assessment is crucial to determine environmental impacts in surrounding areas and the damage that can harm the population. California has collected data on hazardous impacts in different counties in the state. These data compare variables at the population level. This thesis will introduce and emphasize the relationship between environmental science and the hazardous impacts using three techniques: gamma regression, decision trees, and hierarchical modeling. SAS will be used for statistical computing and visualization.

The target variable of the dataset may not always be normally distributed. It is possible that the distribution may be skewed. A gamma regression is fit to a positive response with a right-skewed distribution.

To proceed with variable selection and develop predictive models for the analysis, decision trees are used. Decision trees are either a binary or categorical recursive partitioning method that will produce tree-based models.

The last technique that is used in this analysis is hierarchical modeling. With multilevel models, group level predictors may be included to explore the reasons for the group variations.

Supplemental Files

Some files may require a special program or browser plug-in. More Information

Indexing (document details)
Advisor: Korosteleva, Olga
Commitee: Safer, Alan, Suaray, Kagba
School: California State University, Long Beach
Department: Mathematics and Statistics
School Location: United States -- California
Source: MAI 58/04M(E), Masters Abstracts International
Subjects: Statistics, Environmental science
Publication Number: 10973649
ISBN: 9780438893429
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy