Premium Prediction Based On Grey Fuzzy Markov System Model

Since 1978,when the domestic insurance business gradually resumed,the insurance industry has achieved rapid development,and the domestic premium income is also increasing rapidly year by year.Although China’s insurance industry started late,it has a wide space for future development.As an important indicator to evaluate the development of the insurance industry,premium income has always been the focus of the market.At the same time,the premium income is an important source of profit for each insurance company.Therefore,the forecast of premium income is crucial for the supervision of the entire insurance industry and the decision making of each insurance company,and it can also reflect the development level of the local insurance industry and people’s demand for insurance to some extent.This paper takes the premium income of property insurance and life insurance in Henan province from 2009 to 2018 as the data,taking the grey system theory and the fuzzy markov chain as the research means,and aims at improving the prediction accuracy of the premium,and predicts the premium income of these two kinds of insurance in detail.The main work and conclusions of this paper are as follows:firstly,the grey GM(1,1)and GM(2,1)forecasting models are used to forecast the premium income of property insurance and life insurance in Henan province from2009 to 2018 respectively,and it is found that the grey GM(2,1)forecasting model is more accurate than the grey GM(1,1)forecasting model.Due to weak anti-interference ability of grey prediction model,the knowledge of markov chain is introduced,and use fuzzy set knowledge to fuzzy partition of markov state,thus established the gray GM(2,1)model of fuzzy markov chain.That is,on the basis of the actual value and the predicted value of the grey GM(2,1)prediction model,the residual deviation rate is calculated,and then the residual deviation rate is divided in a fuzzy way,and the grey prediction value obtained before is further modified by using the knowledge of markov chain.Secondly,the gray GM(2,1)fuzzy markov chain model and the greyprediction model are used to predict the property insurance and life insurance premium income of Henan province in 2019.The prediction results show that the model established in this paper can more effectively improve the prediction accuracy than the gray prediction model.Finally,the property insurance and life insurance premium income of Henan province in the next three years is forecasted,and it can be seen that the premium income of Henan province is likely to still show an upward trend,which means that the insurance industry of Henan province may also be developing and making progress in the future.