Modeling eBay auctions
Sometimes modeling a whole dataset suggests trends that disagree with models on separate parts of that dataset. This is known as Simpson's paradox. In the most extreme case, you may see a positive slope on the whole dataset, and negative slopes on every subset of that dataset (or the other way around).
Over the next few exercises, you'll look at eBay auctions of Palm Pilot M515 PDA models.
| variable | meaning |
|---|---|
price |
Final sale price, USD |
openbid |
The opening bid, USD |
auction_type |
How long did the auction last? |
auctions is available as a pandas DataFrame.
This exercise is part of the course
Intermediate Regression with statsmodels in Python
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Take a glimpse at the dataset
print(auctions.info())
# Model price vs. opening bid using auctions
mdl_price_vs_openbid = ____
# See the result
print(mdl_price_vs_openbid.params)