Applying regression models to predict business results

Abstract

In terms of modern business practice, business prediction results are crucially important for evaluation of future financial performance of a company. Planning and prediction procedures are especially important for companies operating under uncertainty. This paper shows an example of planning and prediction of business results in insurance when calculating premium trend by use of linear and nonlinear regression. Due to the uncertainty associated with the moment of claim occurrence and claim amount, it is necessary to secure enough assets to cover the risks. Asset-liability matching requires the prediction of future premium movement per insurance lines which represents the basic concept of development and operation of insurance companies.

Keywords

• linear regression
• nonlinear models
• prediction
• number of policies
• premium