Exploring the Relationship between Environmental Impacts and Different Energy Supplies using Multivariate Time Series

• Main Authors: Jia-XiangLu, 呂嘉祥
• Other Authors: Jeh-Nan Pan
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/43244355724749966765

碩士 === 國立成功大學 === 統計學系碩博士班 === 100 === From the period of Vienna Convention for the Protection of the Ozone Layer(VCPOL) held in 1985 until the Joint United Nations Convention of Climate Change held in 2010, various international organizations started to focus on solving greenhouse gas emission and renewable energy development problems. Recently, the International Standards Organization has launched the latest energy management system ISO50001 in 2011. The purpose of ISO50001 is to establish a standard energy management system and procedure to improve the energy utilization/efficiency and identify its opportunity. The high economic development of human being has resulted in pollution problems and crisis of global ecosystems. Therefore, the reduction of greenhouse gas emissions, increasing the energy efficiency and development of renewable energy have become important research topics for global sustainability. With the increasingly stringent international environmental laws, many leading countries, major organizations and corporations around the world are starting to develop various sustainable energy indices to evaluate their performance on global sustainability. Besides the change in energy utilization structure, establishing an effective energy management system to enhance the green competitive advantages become an ultimate goal of national energy strategy. In this research, we utilize fossil and renewable energy supplies data of Taiwan in the past 29 years (1982 to 2010) to explore the relationship between the impact of environmental key performance indicators of CO2, GDP per capita, fossil and renewable energy by using multivariate time series analysis. The research results show that the prediction accuracy of CO2 emissions by using auto-correlated time series is poor, while the prediction accuracy by using totally fossil energy and renewable energy supplies of two-variable time series model is better than previous one. And the prediction accuracy is the best by using the multivariate time series model, in which includes GDP per capita, coal and renewable energy supplies data are considered. Finally, the trend analysis of CO2 emissions with different growth rates of renewable energy and GDP per capita is performed for evaluating their future environmental impact.