Using multi-linear regression to access the features which best determine "power to weight" of cars The script identifies possible missing values, looks at the relations betwee the different features using their correaltions. Create a new feature "power_to_weight" from the original features and uses the the multi-linear regresson model to forecast the "power_to_weight" values. Furthermore, the performance of the model is evaluated by using the Loss fucnction (mean-squared and root-mean-squared errors) and looking at the explained variance from the model.