Information Technology Reference
In-Depth Information
P2P lending, as a novel economic model, has been studied extensively in recent
years, and is mainly focused on borrowers' social networks and personal information,
loan attributes, lenders' decisions and so on. As for social networks, Freedman & Jin
(2008) have investigated whether they solve the information asymmetry problem in
peer-to-peer lending. They found that loans with friend endorsements and friend bids
have fewer missed payments, but the estimated return of group loans is lower than
those of non-group loans due to lender's learning and the elimination of group
leader rewards. Lin et al. (2009) distinguished between structural and relational
aspects of networks, and found the relational aspects are consistently significant pre-
dictors of the funding probability, interest rates, and ex-post default. Collier and
Hampshire (2010) built a theoretical framework for the evaluation and design of
community reputation systems. Sergio (2009) also built a model-based clustering
method to measures the influence of social interactions in the risk evaluation of a
money request.
To help the lenders make better decision, Luo et al. (2011) proposed a data
driven investment decision-making framework, which exploits the investor composi-
tion of each investment for enhancing decisions making in P2P lending. They re-
vealed that following some investors who have good investment performance in the
past will make more correct investment decisions. Katherine & Sergio (2009) ex-
amined the behavior of lenders and find that, while there exists high variance in risk-
taking between individuals, many transactions represent sub-optimal decisions on the
part of lenders. Klafft (2008) showed that following some simple investment rules
improves profitability of a portfolio. Kumar (2007) empirically proves that
lenders mostly behave rationally and charge appropriate risk premiums for antece-
dents of loan default. Iyer (2009) also find that lenders are able to use available in-
formation to infer a third of the variation in creditworthiness that can be captured by a
borrower's personal information. Puro et al. (2010) developed a borrower decision aid
system, which helps the borrowers quantify their strategic options, such as
starting interest rate, and the amount of loan to request. Wu & Xu (2011)
proposed a decision support system based on intelligent agents in P2P Lending
to help borrowers getting loan more efficiently, by providing borrowers with
individual risk assessment, eligible lender search, lending combination and loan
recommendation.
On Prosper, loan transactions between borrowers and lenders are conducted in an
information-rich environment. When posting a listing, borrowers also submit their
personal portfolios, such as Amount-Requested, Credit-Score, Homeowner, Category
(or purpose), debt information and so on. All these information have influence on
investors' decision. Li & Qiu (2011) displayed that borrower' decisions, e.g., loan
amount, interest rate will determine whether successfully fund loan or not. Herzens-
tein et al (2008) also explored the determinants of funding successfully, found that
borrowers' financial strength and efforts after they post a listing are major factors. The
role of financial intermediaries on the P2P online market was analyzed by Berger &
Gleisner (2009), which demonstrates that the recommendation of a borrower
