The Application Of Monte Carlo Simulation In Designing Trigger Index Of Earthquake Insurance

Earthquake risk gets increasing attention due to the Yushu Earthquake that happened in Qinghai. The methods to provide funds for reconstruction have quickly triggered intense debate in the research institutes. While insurance is considered as the most important source of the funds. Index Insurance, which relies on its mechanism called “a trigger to pay”, has been proved to be a good method to improve the compensation process in catastrophes. Meanwhile, the threat of adverse selection and moral hazard which exposed in traditional insurance can also be avoided.Data is necessarily needed in designing trigger index. However, our country did not keep sufficient record in the past decades. As a result, resolving this issue is the primary requirement to design trigger index. Fortunately, the emergence of Monte Carlo(MC) method provides a reliable basis on solving this problem.Monte Carlo Simulation Method, sometimes called Statistical Simulation or Random Sampling Technique is a very important numerical calculation method based on “random” and the theory of probability and statistics. Monte Carlo Method converses complex research objects or computational problems into random numbers and numerical characteristics for simulation and calculation based on scientific statistical modeling. It simplifies the research object fundamentally, reduces the complexity of computing and leads to the approximate solution with good properties.Currently, MC method has been widely used in the financial sector. Under the inspiration of Caribbean Catastrophe Risk Insurance Facility(CCRIF) and California Earthquake Insurance(CEI), this paper utilizes the MC simulation method to calculate the insurance’s trigger index for earthquakes in Qinghai.The first part elaborates on MC method researches in both domestic and overseas. The chart demonstrates the entire process from data collection to numerical simulation, and then to the bankruptcy probability evaluation. It reveals the relationship between seismic index and MC method. The second part explains the meaning of earthquake index insurance and the index value calculating process. It also introduces MC method’s basic principle and characteristics, which includes convergence, error, applicability, implementation, etc. Next, Easyfit and Matlab is described in the third part, which also introduces their relevant functions. And a model is developed to calculate the trigger point index in the next section which is the empirical analysis part.Qinghai Province is examined carefully in this paper. All required indexes are calculated based on the data of earthquake happened from 1993 to 2016 in this province in excel. The index of each year, along with trigger point I1 and limit point I2 can be selected initially. Then, the optimal distribution of Index Res and Index Com is determined under Easyfit. Meanwhile, random number and inverse function is realized by using Matlab software. Ultimately, by evaluating the bankruptcy probability under different trigger conditions, the thesis gets the first layer 1I =1040, 2I =1488, the second layer 1I ′ =1488, 2I ′ = 1936, the third layer 1I ′ =1936, 2I ′ =2383.73. At last, the historical data is used to verify the rationality of those indexes.