This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License
|
||||||||
|
Paper Details
Paper Title
Credit Risk Forecasting using Deep Learning
Authors
  Shikhar Parikh,  Dr. Jagannath Nirmal
Abstract
Credit risk is the probability of incurring a loss due to a debtor's inability to repay debt. In this paper, we propose to create a Deep Neural Network Classifier (DNN) to forecast the credit risk. We also aim to study the performance of Deep Learning systems for Credit Risk Management on structured data. A comparative study of Credit Risk Forecasting systems using DNN, Logistic Regression, Linear Classifier and Random Forest classification models is conducted. The objective of the classification models is to identify and label the debtors associated with higher credit risk. The DNN Classifier achieves an accuracy of 91.57% and the area under the Receiver Operating Characteristics (ROC) Curve is 89.53%. Since the model is trained on data, which is updated periodically, the model can be utilized to predict future credit risk exposure as well.
Keywords- Area Under Curve, Deep Neural Network, Linear Classifier, Precision, Recall, Receiver Operating Characteristics
Publication Details
Unique Identification Number - IJEDR1903105Page Number(s) - 600-606Pubished in - Volume 7 | Issue 3 | September 2019DOI (Digital Object Identifier) -    Publisher - IJEDR (ISSN - 2321-9939)
Cite this Article
  Shikhar Parikh,  Dr. Jagannath Nirmal,   "Credit Risk Forecasting using Deep Learning", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.7, Issue 3, pp.600-606, September 2019, Available at :http://www.ijedr.org/papers/IJEDR1903105.pdf
Article Preview
|
|
||||||
|