A neural‐network‐based decision‐making model in the peer‐to‐peer lending market

• Author: Shahrooz Bamdad,Golnoosh Babaei
• Published date: 01 July 2020
• Date: 01 July 2020
• DOI: http://doi.org/10.1002/isaf.1480

RESEARCH ARTICLE

A neural-network-based decision-making model in the peer-to-

peer lending market

Golnoosh Babaei|Shahrooz Bamdad

Department of Industrial Engineering, Islamic

Azad University, Tehran, Iran

Correspondence

Shahrooz Bamdad, Department of Industrial

Engineering, Islamic Azad University, South

Tehran Branch. Tehran, Iran.

Email: sh_bamdad@azad.ac.ir

Summary

This study proposes an investment recommendation model for peer-to-peer (P2P)

lending. P2P lenders usually are inexpert, so helping them to make the best decision

for their investments is vital. In this study, while we aim to compare the performance

of different artificial neural network (ANN) models, we evaluate loans from two per-

spectives: risk and return. The net present value (NPV) is considered as the return

variable. To the best of our knowledge, NPV has been used in few studies in the P2P

lending context. Considering the advantages of using NPV, we aim to improve

decision-making models in this market by the use of NPV and the integration of

supervised learning and optimization algorithms that can be considered as one of our

contributions. In order to predict NPV, three ANN models are compared concerning

mean square error, mean absolute error, and root-mean-square error to find the opti-

mal ANN model. Furthermore, for the risk evaluation, the probability of default of

loans is computed using logistic regression. Investors in the P2P lending market can

share their assets between different loans, so the procedure of P2P investment is

similar to portfolio optimization. In this context, we minimize the risk of a portfolio

for a minimum acceptable level of return. To analyse the effectiveness of our pro-

posed model, we compare our decision-making algorithm with the output of a tradi-

tional model. The experimental results on a real-world data set show that our model

leads to a better investment concerning both risk and return.

KEYWORDS

net present value, peer-to-peer lending, portfolio optimization

1|INTRODUCTION

Peer-to-peer (P2P) lending has become popular in recent years

because it reduces financing costs by eliminating the need for a tradi-

tional financial intermediary (Guo, Zhou, Luo, Liu, & Xiong, 2016). In

the P2P lending market, borrowers sign up on online platforms and

submit applications for loans. Lenders (or investors) browse the

available applications, which are called listings, and partially select

some of them for investment (Xia, Liu, & Liu, 2017). If a listing has a

sufficient amount of money, it becomes a loan. Normally, to reduce

the systematic risk, a lender divides the money into different loans,

but most of them are not expert and cannot transform the available

information into smart investments (Mild, Waitz, & Wöckl, 2015).

Therefore, loan evaluation is an effective tool to guide unprofessional

lenders in making rational decisions (Guo et al., 2016). As making deci-

sions on whether or not to fund a certain loan is challenging for

lenders in the P2P lending market, various loan evaluation models,

namely random forest (Malekipirbazari & Aksakalli, 2015), decision

tree (Serrano-Cinca & Gutiérrez-Nieto, 2016), artificial neural network

(Byanjankar, Heikkila, & Mezei, 2015), and kernel method (Guo

et al., 2016), have been established.

The P2P lending market is new and developing, so research on

the loan evaluation models for P2P lending is limited. In 2015, a

random-forest-based classification method was proposed by

Received: 7 January 2020Revised: 2 April 2020Accepted: 19 June 2020

DOI: 10.1002/isaf.1480

142 © 2020 John Wiley & Sons, Ltd.Intell Sys Acc Fin Mgmt. 2020;27:142–150.wileyonlinelibrary.com/journal/isaf