暴露在空氣汙染的環境中將導致許多疾病產生，因此空氣汙染儼然成為各國迫切解決的問題之一。台灣也不遺餘力地在空氣汙染防制的工作上做出許多努力，而比較台灣各地區空氣汙染程度發現，即使市政府每年皆祭出許多空氣汙染改善政策，高雄市歷年排名一直位處最嚴重城市，民眾對空氣品質也仍感受不到改善，因此本研究所欲探討市政府致力於改善空氣品質是否有效。
本文以高雄市2011年1月至2018年12月的月資料做為研究樣本，針對燃油汽機車數目、污染防治歲出金額、公共運輸接駁人數、臭氧及PM2.5作為分析變數。首先利用單根檢定來確認所有變數透過差分後為定態，再使用共整合檢定分析變數間是否存有長期關係，若有則建立向量誤差修正模型；若無則建立向量自我迴歸模型，藉此觀察空氣汙染物與各個變數的落後期之間的相關性，最後再利用格蘭傑因果關係檢定觀察變數間是否存在因果關係。結果顯示，污染防治歲出金額與臭氧存在格蘭傑因果關係；公共運輸接駁人數與PM2.5存在格蘭傑因果關係；燃油汽機車數目與臭氧、細懸浮微粒存在格蘭傑因果關係。代表統計上高雄市於2011年至2018年間的空氣品質有得到改善。

Exposure to air pollution will lead to many diseases, so air pollution has become one of the imperative problems for all countries. Taiwan has spared no effort to make a lot of efforts in air pollution prevention and control. Compared with the level of air pollution in various parts of Taiwan, it is found that even if the municipal government provides many air pollution improvement policies every year, Kaohsiung City has been ranked the most serious city in the past years. The public still does not feel the improvement of air quality, so the study wants to explore whether the municipal government is committed to improving air quality.
This paper takes the monthly data of Kaohsiung City from January 2011 to December 2018 as a research sample, and analyzes the number of fuel turbines, the amount of pollution prevention and control, the number of public transport connection, ozone and PM2.5. First, use a single root check to confirm that all variables pass the first-order difference and then determine the state. Then, if there is a long-term relationship between the variables, the vector error correction model is established. If not, the vector autoregression model is established. Observe the correlation between air pollutants and the backward period of each variable. Finally, use Granger causality test to see if there is a causal relationship between variables. The results show that there is Granger causality between the amount of pollution prevention and the ozone; the Granger causal relationship between the number of public transport connection and PM2.5; the number of fuel turbines and the Granger causal relationship between ozone and PM2.5. On behalf of the statistics, Kaohsiung City's air quality have improved between 2011 and 2018.

論文審定書 ……………………………………………………………………… i
摘要 …………………………………………………………... …….…................ ii
Abstrict ….……………………………………..………………………………… iii
第 一 章 緒論……..……………………………………………………………. 1
1.1 空氣污染的嚴重性………………………………………………………… 1
1.2 台灣空氣品質現況……………………………………………………….... 2
1.3 研究目的………………………………………………………………….. 11
第 二 章 文獻回顧…………………………………………………………... 12
2.1 探討不同研究方法之文獻………………………………..........…………. 12
2.2 探討交通及工業汙染之文獻……………………………........................ 14
2.3 探討不同經濟發展程度之文獻………………………………............... 15
第 三 章 研究方法……………………………………………………………. 17
3.1 單根檢定………………………………………………………………… 17
3.2 共整合檢定……………………………………………………………… 21
3.3 誤差修正模型…………………………………………………………… 24
3.4 向量自我迴歸.………………………………………………………...… 25
3.5 格蘭傑因果關係檢定…………………………………………………… 27
第 四 章 實證結果…………………………………………………………… 29
4.1 資料說明...……………………………………………………………...... 29
4.2 單根檢定結果………………………………………………………..….... 30
4.3 共整合檢定結果………………………………………………………..... 31
4.4 向量誤差修正模型結果…………………………..…………………….... 32
4.5 格蘭傑因果關係檢定結果……………………..……………………….... 34
第 五 章 結論及建議……………………………………………………….. 36
5.1 結論……………………………………..................................................... 36
5.2 未來研究建議……………………………………...................................... 37
參考文獻…………………………………………………………….….………… 38
 
圖次
圖1-1 台灣106全年度PM10排放總量來源圖………………………………..……6
圖1-2 台灣106全年度PM2.5排放總量來源圖………………………………….…7
圖1-3 台灣106全年度 SO_x 排放總量來源圖…………………………………….…7
圖1-4 台灣106全年度 NO_x 排放總量來源圖…………………………………….…8
圖1-5 台灣106全年度CO排放總量來源圖……………………………………..….8
圖1-6 六都空氣汙染達不良天數佔全年比例…………………………………...…10

表次
表1-1 PSI下污染物濃度與副指標值對照表…………………………………..….…2
表1-2 PSI下空氣汙染與健康影響對照表………………………………………...…3
表1-3 PM2.5指標換算表與活動建議……………………………………………..…4
表1-4 AQI下空氣汙染與健康影響對照圖………………..…………………………5
表1-5 台灣106全年度汙染排放主要來源圖……………………………………..…6
表1-6 107年度AQI指標汙染物不良天數比例……………………….……………9
表4-1 變數說明表…………………………………………………………………...29
表4-2 ADF單根檢定表……………………………………………………...………29
表4-3 最適落後期數選取結果……………………………………………….……..30
表4-4 Johansen共整合檢定結果……………………………………………...…….31
表4-5 誤差修正模型結果…………………………………………….……………..32
表4-6 Granger因果關係檢定結果………………………………………….……….33

一、中文部分

1. 陳旭昇(2013)，「時間序列分析：總體經濟與財務金融之應用」，台北市，台灣東華
2. 張立農(2014)，交通及工業區空氣品質監測站PM10濃度影響因素之研究，國立中興大學水土保持學系研究所論文
3. 黃俊憲(2014)，台中都會區機動車輛排放減量對臭氧生成量之影響，中興大學環境工程學系碩士學位論文
4. 黃小明、范雪敏、李治(2016)，武漢市PM10和PM2.5的分析及預測，華中農業大學理學院
5. 黃彥齊(2018)，類神經網路應用於空氣品質預測與異常偵測之研究，國立交通大學環境工程研究所碩士論文
6. 楊奕農(2009)，「時間序列分析 經濟與財務上之應用」，台北市，雙葉書廊
7. 鍾惠民、周賓鳳、孫而音(2014)，「財務計量Eviews的運用」，台北市，新陸書局

二、英文部分

1. Athanasios Sfetsos and Diamando Vlachogiannis. (2013). “An analysis of ozone variation in the Greater Athens Area using Granger Causality.” Atmospheric Pollution Research 4, pp.290-297.

2. Dean E. Schraufnagel, John R. Balmes, Clayton T. Cowl, Sara De Matteis, Soon-Hee Jung, Kevin Mortimer, Rogelio Perez-Padilla, Mary B. Rice, Horacio Riojas-Rodriguez, Akshay Sood, George D. Thurston, Teresa ToAnessa Vanker, and Donald J. Wuebbles. (2019). “Air Pollution and Noncommunicable Diseases.” Chest Journal.155(2), pp. 409-416.

3. Dipankor Coondoo and Soumyananda Dinda. (2002). “Causality between income and emission: a country group-specific econometric analysis.” Ecological Economics. 40, pp. 351–367.

4. Haoran Zhao, Sen Guo and Huiru Zhao. (2018). “Characterizing the Influences of Economic Development, Energy Consumption, Urbanization, Industrialization and Vehicles Amount on PM2.5 Concentrations of China.” Sustainability 10, pp. 2574.

5. Jing Chen, Chunshan Zhou, Shaojian Wang and Shijie Li. (2018). “Impacts of energy consumption structure, energy intensity, economic growth, urbanization on PM2.5 concentrations in countries globally.” Applied Energy 230, pp. 94-105.

6. Lu Liu, Chong Zhou, Junbing Huang and Yu Hao. (2018). “The impact of Financial Development on Energy Demand: Evidence from China.” Emerging Markets Finance and Trade 54, pp. 269-287.

7. Mohammad Jahangir Alam , Ismat Ara Begum , Jeroen Buyssed , Sanzidur Rahman and Guido Van Huylenbroeck. (2011). “Dynamic modeling of causal relationship between energy consumption, CO2 emissions and economic growth in India.” Renewable and Sustainable Energy Reviews. 15, pp. 3243– 3251.

8. Mohammad Jahangir Alam, Ismat Ara Begum , Jeroen Buysse and Guido Van Huylenbroeck. (2012). “Energy consumption, carbon emissions and economic growth nexus in Bangladesh: Cointegration and dynamic causality analysis.” Energy Policy. 45, pp. 217–225.

9. Muhammad Shahbaz, Muhammad Zeshan and Talat Afza. (2012). “Is energy consumption effective to spur economic growth in Pakistan? New evidence from bounds test to level relationships and Granger causality tests.” Economic Modelling 29, pp. 2310–2319.

9. Usama Al-Mulali, Ilhan Ozturk and Hooi Hooi Lean. (2015). “The influence of economic growth, urbanization, trade openness, financial development, and renewable energy on pollution in Europe.” Natural Hazards 79, pp. 621-644.

11. Yu Hao and Yi-Ming Liu. (2016). “The influential factors of urban PM2.5 concentrations in China: a spatial econometric analysis.” Journal of Cleaner Production 112, pp.1443-1453.
 
三、網站資料

1. 行政院環保署空氣品質監測網
網址：https://taqm.epa.gov.tw/taqm/tw/default.aspx
2. 高雄市環境保護局
網址：https://www.ksepb.gov.tw/Home/Index?disktop=true
3. 高雄市空氣汙染防制網站
網址：https://air.ksepb.gov.tw/
4. 重要市政統計資料庫查詢網址
網址：http://pxweb.kcg.gov.tw/pxweb2007p/dialog/treelist/folder.htm
5. 中華民國統計資訊網
網址：https://statdb.dgbas.gov.tw/pxweb/Dialog/statfile9.asp
6. 全國空氣汙染物排放量清冊資訊系統
網址：https://teds.epa.gov.tw/
7. 行政院環保署空氣品質監測年報
網址：https://www.epa.gov.tw/Page/672FA2BDDEAA22C7/71fedbd8-9829-49e0
-b02b-0addb5bd470a