Data used here came from the Darden Case Study: "Advanced Insurance". 

Working with your learning team, using the training data , generative AI resources of your choice, and the visualizations we explored in class, build the best model you can to predict the missing premiums in the Advanced Test tab . Submit your predictions for the missing premiums by 11:00 PM on Tuesday (Feb 17). You should come to our final class prepared to discuss your model and approach.

When you submit a set of predictions, you will see where your teams ends up on the leaderboard and get feedback on your model in the form of the RMSE (root mean squared error). The smaller the error, the better!

You are welcome to use ANY technology to complete this task. In fact, non-excel tech is encouraged -- have fun with it!  Please be prepared to share if/how generative AI was helpful. If you use non-excel tools for your analysis (e.g., python, R, etc.), we would love for you to demonstrate your approach in class. (All such tools are enrichment and, for the avoidance of doubt, will not be necessary on the exam).

**File descriptions**
- training.csv - the training set of 7500 rows and 18 columns
- test.csv - the test set of 1500 rows and 18 columns. The column of 'Comp_A_Premium' has no values
- Submission.csv - a sample submission file in the correct format

**Data fields**
![](https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F228069%2F28b99133a267d31c702051056eadc905%2FScreenshot%202026-02-08%20175823.gif?generation=1770591523538523&alt=media)
- ID: A unique customer ID to identify accounts
- Comp_A_Premium:  The annual rate that Competitor A charges to insure one vehicle for one year
- SafeDrive_Score: The output of Advanced Insurance's in-house prediction model to assess the safety of a driver. Higher scores indicate a safer driver
- Age: 	The age of the insured driver
- Years_Driving:  The number of years of driving experience of the insured driver
- Gender	The reported gender of the insured driver
- Vehicle_Type	A code for the type of vehicle insured
- SC_or_MC?	An indication of whether the insured driver purchases insurance for a single vehicle (SC) or multiple vehicles (MC)
- Points	The number of penalty points on the insured driver's record. Higher points indicate more frequent or severe driving violations. 
- County_Residence	A categorical variable indicating the county of residence of the insured driver
- PIF?	A categorical variable indicating whether the insured driver's account with Competitor A has been paid in full. (1= paid in full)
- Homowner	A categorical variable indicated whether the insured driver is a homeowner (1 = homeowner)
- Education	A categorical variable indicating the highest level of education achieved by the insured driver.
- Marital_Status	A categorical variable indicating whether the insured driver is married (1 = married)
-  Organ_Donor?	A categorical variable indicating whether the insured driver is a registered organ doner (1 = organ doner)
- Months_Since_Policy_Inception	The number of months that have elapsed since the insured driver initiated a policy with Competitor A
- Log(SafeDrive_Score)	The natural logarithm of Advance Insurance's in-house prediction model to assess the safety of a driver