August 11, 2012 – So how does the Lending Club Algorithm work? (Lending Club Strategy)

Here at P2P Analytics, we’ve received some emails asking the following:

1. What strategy does the Lending Club selection algorithm implement to rank notes?
2. And what exactly is the score that we see in the distribution. What does this score signify?

To answer the first question:

In constructing the algorithm to estimate default rate probability, the first step was to find the top 12 factors that corresponded with influencing default rates:

Grade
Housing status (Home, Mortgage, Rent)
Job time length
Loan Term
Borrower Text description (and whether it was entered)
Total Credit Lines
Income
Type of Loan (Consolidation, Green, Car, etc)
Earliest Credit Line
Number of Credit Inquiries
DTI ratio
Amount Requested
Months since last delinquency

The second step was to check the effect of each of these variables across each grade. For example, Housing status was checked for it’s effect on default rates across all A,B,C, etc loans and a table was built of these variables that described how much each factor influenced default rates (for better of for worse). Each time that variable was checked across a grade, all of the notes ever issued (except the very recent notes) in that grade to make the determination.

A few things to keep in mind about the Lending Club algorithm:

Each variable was built across all loans ever issued for each grade. In other words, thousands of samples were used to make a determination as to the correlation of each variable. Due to the fact that there are twelve correlating factors used, this increases the accuracy of the algorithm’s determinations. Imagine if the algorithm only took into account 3 or 4 factors. It would probably be a better strategy than just guessing when selecting notes, but it still would be a very coarse selection method.

To answer the second question, the score is the output of the algorithm. It’s a number that is a rough estimation of the default rate probability of a particular note. The lower this number is, the better. As the algorithm currently stands, more tweaking needs to be done to make this score slightly more representative in terms of matching actual default rates, but overall the algorithm is very good at comparing notes in funding.

You can leave a response, or trackback from your own site.

4 Responses to “August 11, 2012 – So how does the Lending Club Algorithm work? (Lending Club Strategy)”

  1. Conor says:

    Are the 12 factors analysed independently or collectively against all notes?

    We may see mortgages perform better than renters overall, and 36-month perform better than 60-month overall as well. However, together, 60-month and mortgage may perform the best of all.

  2. jasonejacks says:

    Yes, all factors are analyzed independently. The algorithm does exactly what you mention: A note that has two independently favorable factors (such as mortgage + 60 month term) does yield a better rank than a note with just one of the factors.

  3. Conor says:

    My point is that it may not always hold true that independent analysis justifies the best collective results. You could see underperforming independent factors perform better when combined with other factors. In my example, I said 36-month loans were performing better than 60-month loans if measured independently (I’m not sure if this is true, just an example), but that 60-month loans could perform better than 36-month loans if only looking at loans with people with a mortgage. In that case, shouldn’t a note with mortgage + 60 month yield better, even though 36-month is better when looked at independently?

  4. Roundup of Social Lending News – August 18, 2012 says:

    [...] P2P Analytics - SO HOW DOES THE LENDING CLUB ALGORITHM WORK? (LENDING CLUB STRATEGY) – Social Lending.net[...]

Leave a Reply


3 + = six

Note: The material presented on P2Panalytics is provided for informational purposes only and is based upon information that is considered to be reliable. However, neither P2PAnalytics nor its affiliates warrant its completeness, accuracy or adequacy and it should not be relied upon as such. Neither P2PAnalytics nor its affiliates are responsible for any errors or omissions or for results obtained from the use of this information. Past performance, including the tracking of Peer to Peer Lending notes and portfolios for educational purposes, is not necessarily indicative of future results. Anyone related to P2PAnalytics may hold positions in the notes mentioned, which may change at any time without notice. Peer to Peer lending has inherent risks and P2PAnalytics is not responsible for investment losses incurred by any of its users or readers.

This material is not intended as an offer or solicitation for the purchase or sale of any security or other financial instrument. Securities or other financial instruments mentioned in this material are not suitable for all investors. Any opinions expressed herein are given in good faith, are subject to change without notice, and are only intended at the moment of their issue as conditions quickly change. The information contained herein does not constitute advice on the tax consequences of making any particular investment decision. This material does not take into account your particular investment objectives, financial situations or needs and is not intended as a recommendation to you of any particular securities, financial instruments or strategies. Before investing, you should consider whether it is suitable for your particular circumstances and, as necessary, seek professional advice.