Repository for performing Data exploration, several statistical tests, regression analysis , variable selection, feature engineering and XGBoost model to predict loss of an insurance claim.
Use of statistical machine learning for solving many real-world problems across several domains is so prominent as well as very useful.Insurance domain is one such area, where machine learning can be used to predict ‘loss’ of an insurance claim by using a set of parameters.
Our goal for this project is to predict the (response variable) ‘loss’ of an claim based on several independent variables. These independent variables includes both continuous and categorical predictor variables, Mean Absolute Error(MAE) is the error metric that we will be minimising during the training process of several models.