Latest Comments
« Combining Multiple Model Probabilities | Main | Mental health is important »

Prosper P2P Loan Model

My latest side project has been working on modeling Prosper P2P loans. Features of the data like credit grade, income, etc. are all readily available for import (it's like they actually want you to use this data!). All of the data is imported in MySQL using a script, and then sliced and diced using R.

I created a runtime model, which is an ensemble of various machine learning models (think Netflix prize grand prize solution). The runtime model runs on a Raspberry Pi and uses the Prosper API to place bids automatically.

Reader Comments (2)

I'm thinking of doing something like this... I came across your post by googling 'machine learning prosper data.' I'm wondering how it's been going for you?

July 19, 2013 | Unregistered CommenterD

So far, after 7 months of algorithmic trading, I have ~150 loans funded by the algorithm at an money-weighted average interest rate of 21.7%. There are one >15 days late loan and one <15 days late loan. I am waiting until I have some notes that are considered "seasoned" at 10 months old before I write an update, but overall, I am pretty happy considering that I already have three manually-curated loans default on me!

August 10, 2013 | Registered CommenterYew Jin

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>