Evaluation and Prediction Model for Exploring the Impact of Climate Change on National Vulnerability

Abstract

Climate change’s pervasive influence on human life has been an inevitable topic nowadays, especially the impacts on instability of a state has been attached with great importance recently. In this paper, we establish evaluation model and regression model to complete the missions. First, based on the scheme of Fragile State Indexwhose weights of 12 indicators are uncertain, we are aimed to decide them to complete the Evaluation Model. Fuzzy Analytic Hierarchy Process and Entropy Weight Method are adopted to combine the advantages of human knowledge and data pattern. Then, in order to find out how climate change affects the 12 indicators, Regression Model is established. We firstly identify 5 indicators including disaster, water, land, temperature and CO₂ to describe climate change. After that, we identify 6 control variables to reduce system error. Last but not the least, taking the difference between countries, we use fixed cross-section effects model to regress. Data of 115 countries from 2009 to 2019 is collected and all the R2 of the 12 regression models are above 0.9, indicating a good fitting result.Finally, we establish a Forecast Model to evaluate the actual effect of our work. Based on collected climate data, Time Series ARIMA Model is adopted.