Ambient pollutant, especially ground level ozone that causes respiratory diseases, has been a great concern in Southern California. U.S. Environmental Protection Agency provides the Air Quality Index (AQI) as a tool to assist the public of health warnings. AQI for ozone is currently divided into six states depending on the level of public health concern. In statistical point of view AQI can be characterized as nonstationary ordinal-valued time series. The purpose of this study is to implement statistical models for short-term forecasting of AQI. This thesis presents a generalized linear type modeling to handle the autocorrelated ordinal time series. The model is applied with four different link functions: identity, logit, probit, and complementary log-log and their forecast performance are compared. Random time-varying covariates include past AQI state, various meteorological processes, and periodic component. Data used in this study are AQI for ozone from five monitoring stations in San Bernardino County, CA for 2004 to 2006.
For the purpose of evaluating the performance of one-day-ahead forecast, the 2007 data from the same place are used. The meteorological data are from the nearby Barstow city in San Bernardino County. The portmanteau test is used to test error autocorrelations. The partial likelihood ratio test, Akaike information criterion (AIC), and Bayesian information criterion (BIC) are used to measure the goodness of fit and compare the models. The results show the model well captures the nonstationarity in ozone process and remove the nonstationarity in residuals. Both logit and probit models correctly forecast about 85% of the observed AQI.
|Advisor:||Kim, Sung Eun|
|School:||California State University, Long Beach|
|School Location:||United States -- California|
|Source:||MAI 51/01M(E), Masters Abstracts International|
|Subjects:||Statistics, Atmospheric sciences, Environmental science|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be