Forecasting the Cost of Insurance Using Machine Learning Methods

Predicting insurance costs and associated payouts is a key task in the insurance sector, impacting risk management and financial stability of companies. Traditional approaches primarily rely on statistical methods and supervised learning, yet they often fail to account for the dynamics and volatility of the insurance market, leading to inaccurate forecasts. The potential of machine learning methods, including regression models and ensemble techniques, to enhance the accuracy of predicting insurance policy costs and payouts is significant. The aim of this study is to analyze data from insurance companies regarding medical insurance costs and payouts, to explore the relationships between various factors, and to develop and compare regression machine learning models for predicting these economic indicators. The research involves data collection and processing, analysis of relationships, and model construction. The outcome will include recommendations for selecting the most suitable models and configuring their hyperparameters to increase accuracy and stability of predictions under market variability. The implementation of these approaches will allow insurance companies to manage risks more effectively and increase financial stability.