Default Risk Prediction Based on Support Vector Machine and Logit Support Vector Machine

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216275%3A25410%2F23%3A39920854" target="_blank" >RIV/00216275:25410/23:39920854 - isvavai.cz</a>

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-031-18552-6_6" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-031-18552-6_6</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-18552-6_6" target="_blank" >10.1007/978-3-031-18552-6_6</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Default Risk Prediction Based on Support Vector Machine and Logit Support Vector Machine

Popis výsledku v původním jazyce

This chapter aims to predict the credit customer default risk. We propose a machine learning algorithm such as Support Vector Machine and a hybrid default risk prediction model such as Logistic Regression and Support Vector Machine being known as LogitSVM (LSVM) to access the credit default risk. We apply three real-world credit databases to validate the probability and value of the proposed risk appraisal hybrid approaches. This chapter uses Type-I Error, Type-II Error, and Root Mean Squared Error (RMSE) to evaluate the performance of the algorithms. Empirical findings show that hybrid model experimentation (LogitSVM) maximizes overall accuracy and minimizes RMSE, Type-I error, and Type-II error. This study is useful for stakeholders to develop a wide variety of approaches to predict risk of default of the credit customer.

Název v anglickém jazyce

Default Risk Prediction Based on Support Vector Machine and Logit Support Vector Machine

Popis výsledku anglicky

This chapter aims to predict the credit customer default risk. We propose a machine learning algorithm such as Support Vector Machine and a hybrid default risk prediction model such as Logistic Regression and Support Vector Machine being known as LogitSVM (LSVM) to access the credit default risk. We apply three real-world credit databases to validate the probability and value of the proposed risk appraisal hybrid approaches. This chapter uses Type-I Error, Type-II Error, and Root Mean Squared Error (RMSE) to evaluate the performance of the algorithms. Empirical findings show that hybrid model experimentation (LogitSVM) maximizes overall accuracy and minimizes RMSE, Type-I error, and Type-II error. This study is useful for stakeholders to develop a wide variety of approaches to predict risk of default of the credit customer.

Druh

C - Kapitola v odborné knize

CEP obor

—

OECD FORD obor

50206 - Finance

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2023

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název knihy nebo sborníku

Novel financial applications of machine learning and deep learning : algorithms, product modeling, and applications

ISBN

978-3-031-18551-9

Počet stran výsledku

14

Strana od-do

93-106

Počet stran knihy

231

Název nakladatele

Springer Nature Switzerland AG

Místo vydání

Cham

Kód UT WoS kapitoly

—