A Study on the Decision of Insurance Order Quantity in Disaster-prone Areas Based on Grey Prediction and Multivariate Linear Regression

In recent years, earthquakes, floods, sandstorms, extreme heat and other natural disasters have occurred with increasing frequency in all corners of the Earth, and these disasters have caused huge economic losses. In this paper, the data of the disaster-prone state of Florida is taken as an example. In order to make the local insurance companies better make insurance decisions, the article first estimate the disaster frequency of the region in the next five years based on Grey Prediction, and then obtain the average expected weather-related disaster frequency of Sky and Land disasters (431.24 and 428.86). At the same time, the article carries out Multivariate Linear Regression on insurance sales (dependent variable) and disaster frequency (independent variable), obtaining the linear regression equation of the state, and calculates the expected value of insurance sales in the next five years (14.2372 million). Then, paper uses the concept of marginal benefit in economics and curve fitting toolbox to fit the marginal insurance cost curve and marginal insurance revenue curve, and take the intersection point of the curve (20.2692 million) as the number of underwriting orders that make the insurance company most profitable. Through comparison, it is found that the underwriters in Florida should continue to conduct business.