The analysis is conducted to help Drivetime, the second largest vehicle retailer in the US, to select the “right” vehicles for their inventory. In the used vehicle market, normal dealer sales occur within 90 days of the delivery to the dealership. If the vehicle does not sell within the 90 days, it’s called an overage vehicle. Overage vehicle will be sold at lower price than normal, incurring a loss in the profit for Drivetime.
The project aims to provide a thorough understanding of the factors that drive the time a used vehicle staying in the dealer lot before being sold. Furthermore, it can be used to guide the selection process of Drivetime, helping them to have more normal sales, avoid overage and maximize their profit.
This project is inspired by one of the case studies in the book “Marketing Data Science: Modeling Techniques in Predictive Analytics with R and Python.” by Thomas Miller. The author provided the dataset and problem statement.
The analysis contains 3 main parts:
- Exploratory Data Analysis: understanding the dataset
- Model Selection & Comparison: We will compare between 2 models and examine the effect of using statistical model in selection process on Drivetime’s profit
- Recommendation - value added part to Drivetime
The folder contains:
- Dataset of used vehicles by Drivetime (drive_time_sedans.csv)
- EDA and Modeling (main.ipynb)
- Final Report (Report.docx)