The 2nd session was the Big Lender Panel. It was great. Worth the price of admission. The Lenders (with links to their portfolio) were pensioner, L5, and janeybooboo. They are all manual bidders and from the sound of it they spend quite a bit of time vetting the borrowers. The biggest thing I took away from it is that they have no problem filling their capacity. The question I would like to ask them that I didn’t is this….
Subtracting the default rate from the interest rate does not give the ROI. (I think they were implying it was an approximation for ROI.) In order to calculate ROI you need to use a Markov model and assume some default characteristics such as default rates over the life of the loan. (Similar to the new performance tool, but I think they are making some assumptions.) Given that I use standing orders what I would rather do is Monte Carlo a Standing Order against the data and see the actual IRR. Are they in fact using a Markov model or similar? And I would love to know what their Prosper IRR is using my definition.
Update L5 responds: http://www.rateladder.com/2007/02/14/avg-rate-minus-default-rate-does-not-equal-expected-return/
Update pensioner responds: http://www.rateladder.com/2007/02/14/pensioner-responds-avg-rate-minus-default-rate-does-not-equal-expected-return/
The morning session (Lender Portfolio Analysis) was kinda blah and I will spare the details… I found it to be a rehash of the forums advice. The new ROI tool was one topic. I learned a little about it’s internals, but I look forward to reading the “white paper”. Here is a link: http://www.prosper.com/public/help/topics/lender-marketplace_performance_calculation.aspx
If you liked this article, vote for it on del.icio.us and stumbleupon.