The GLS and GAM Time Series Model for the Mortality and Environment Factors of Taiwan
| Autor: | Wei-Hsiu Lin, 林韋秀 |
|---|---|
| Rok vydání: | 2007 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 95 In recent years, the air pollution problem becomes more and more serious. People can’t live without air, so human life is threatened by air pollution. According to the study, serious air pollution situation will result in deadly influence to people. Nowadays, Taiwan is facing this crucial. Taiwan is located in semi tropics area, so temperature does not change much. The low temperature causes deadly influence in Taiwan during the cold current. So we want to analyze the impact on mortality about the factors of air pollution and temperature. Since there are different conditions in different area, we will discuss these circumstances about Taipei, Taichung, Kaohsiung and the whole Taiwan Island. Because these environment factors are time related, we should analyze the sdtudy by using time series method, therefore Generalized Least Square (GLS) and Generalized Additive Model (GAM) are adopted in this research. About GLS, we use time series model auto-regressive moving-average process (ARMA) to fit time related residuals. About GAM, we fit time related property by using smooth function. So there will be differences between GLS and GAM. Both of GLS and GAM are cotaining important factors. Besides, GAM will contain more factors than GLS. After analyzing, we have some findings. In different area, different environmental factors will result in different levels impacts with the same cause of death. On the other hand, in the same area, different environmental factors will result in different levels impacts with different cause of death. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|