There are some challenging questions on the last page, anyone has the correct answers to them? I have no clue to start the first question. Thank you in advance.
Analysis next steps:
- Find the most common brand/model combinations
- Split the
odometer_km into groups, and use aggregation to see if average prices follows any patterns based on the milage.
- How much cheaper are cars with damage than their non-damaged counterparts?
I will try to add some details to these steps so that you can attempt them on your own first!
most common brand/ model. Just like how you did the Top brand by mean price, try to identify the car model for each of the brands that sell more. Meaning each brand must have 5-6 models associated with it. We want to analyze out of these models, which is the Top selling for that specific brand.
So per brand, there would be one model. It would be interesting to see if there’s a tie!
split the odometer… create bins for odometer readings then calculate average prices for these bins. I chose 4 bins and I don’t know why I left one particular interval out of the calculation altogether So my calculation should be wrong here. Nevertheless, try to identify the pattern/ correlation if any between average prices and bins.
damage vs non-damage… Try to segregate data into 2 parts - cars with damage and without. Analyze the avg selling price for them; this is to understand if damage effects the sale of cars (its but obvious that damaged cars would sell for cheap … but confirm it via data)
Let’s keep the Correct answer aside for now. I am attaching here my solution notebook. By the way, I did it when I was also a newbie to everything “DS with Python” (and we still have loads to learn)!
So the project lacks everything that I usually mention in other students’ project feedback and also technically it should be pretty lame.
Scroll down to Section 9!
Other students have also submitted their project with these additional questions, so you can check out their work as well to get a better understanding of these questions.
ExploreEbayCarSalesData.ipynb (79.0 KB)
Click here to view the jupyter notebook file in a new tab