Using predictive modeling in insurance underwriting Research Paper

Using predictive modeling in insurance underwriting - Research Paper Example Underwriting is done by ensuring that there is more collection of premium and investment income by the insurance company than the losses paid out by the company. This can be done by implementing new strategies of pricing of premium to the customers. The below formula should be used to ensure that there is profit realization by the insurance company. The risk of the damaged is transferred from the insurer to the insurance company by compensating the damage. The compensation of the damage may be higher than that of the charges from the insurer. In case there is such occurrence, the insurance companies, process the clients demand by forming mitigate and consortium risk amongst them (Siegel, 2013). For example, the company tries to balance the various types of insurance hazards such as fire to cater for other costs incurred. Pricing to cater for different risk involves the comparison between the actual losses with relative losses. In this method, multivariate or univariate analyses and probability analyses are used to approximate ratemaking and approximating the rate of future claims based on a given risk. Therefore, the decision on premium to be charged is a major problem in the insurance companies. As stated above, the claims and the underwritten cost should be less compared to the premium received by the company (Siegel, 2013). The K- means clustering can be used to determine or classify the customer based on the risk, the characteristics of a customer, and also determining the probability of the customer based on their characteristics. Through this method, the company can target a certain group of customers by different marketing plans and increase the profitability of the company. The use of clustering method can help in developing new products or cross selling product by understanding the market behavior (Zhu & Davidson, 2007). For example, some of the customers can fall under two or more covers at the same time due to different demographic