submit idea 1 - #31
Conversation
There was a problem hiding this comment.
You could run a regression using certain housing statistics to see which ones are most significant in predicting the prices of homes, while accounting for locations. Maybe there are home features that are more impactful in the selling process on the west coast versus the east coast of the US, just a thought.
|
I remember talking with you about your idea. I think this is great. One thing to note is you should probably figure out what city you want to work with. It's probably good to pick one city and look at different neighborhoods. Once you scrape the website, you'd have a lot of cool data to explore. Price prediction models make a lot of sense. Because of real estate being considered as a financial instrument as a secondary and sometimes primary consideration, make sure to look at asset pricing models from finance. |
There was a problem hiding this comment.
Interesting idea! It seems like the two datasets you are comparing need to have a common column, would that be the date? You will need to create some thresholds so that the model will know whether each observation should be categorized/labeled as positive, negative, or neutral.
|
3 participants
No description provided.