Business Distress Prediction Using Bayesian Logistic Model for Indian Firms

Arvind Shrivastava, Kuldeep Kumar, Nitin Kumar

Research output: Contribution to journalArticleResearchpeer-review

40 Downloads (Pure)

Abstract

The objective of the study is to perform corporate distress prediction for an emerging economy, such as India, where bankruptcy details of firms are not available. Exhaustive panel dataset extracted from Capital IQ has been employed for the purpose. Foremost, the study contributes by devising novel framework to capture incipient signs of distress for Indian firms by employing a combination of firm specific parameters. The strategy not only enables enlarging the sample of distressed firms but also enables to obtain robust results. The analysis applies both standard Logistic and Bayesian modeling to predict distressed firms in Indian corporate sector. Thereby, a comparison of predictive ability of the two approaches has been carried out. Both in-sample and out of sample evaluation reveal a consistently better predictive capability employing Bayesian methodology. The study provides useful structure to indicate the early signals of failure in Indian corporate sector that is otherwise limited in literature.

Original languageEnglish
Article number113
Number of pages15
JournalRisks
Volume6
Issue number4
DOIs
Publication statusPublished - 9 Oct 2018

Fingerprint

Distress
Logistic model
Prediction
Business sector
Bayesian modeling
Methodology
Logistics
Evaluation
Emerging economies
Bankruptcy
India
Predictive ability

Cite this

Shrivastava, Arvind ; Kumar, Kuldeep ; Kumar, Nitin. / Business Distress Prediction Using Bayesian Logistic Model for Indian Firms. In: Risks. 2018 ; Vol. 6, No. 4.
@article{c17307ac07f6407a8177a011b061f219,
title = "Business Distress Prediction Using Bayesian Logistic Model for Indian Firms",
abstract = "The objective of the study is to perform corporate distress prediction for an emerging economy, such as India, where bankruptcy details of firms are not available. Exhaustive panel dataset extracted from Capital IQ has been employed for the purpose. Foremost, the study contributes by devising novel framework to capture incipient signs of distress for Indian firms by employing a combination of firm specific parameters. The strategy not only enables enlarging the sample of distressed firms but also enables to obtain robust results. The analysis applies both standard Logistic and Bayesian modeling to predict distressed firms in Indian corporate sector. Thereby, a comparison of predictive ability of the two approaches has been carried out. Both in-sample and out of sample evaluation reveal a consistently better predictive capability employing Bayesian methodology. The study provides useful structure to indicate the early signals of failure in Indian corporate sector that is otherwise limited in literature.",
author = "Arvind Shrivastava and Kuldeep Kumar and Nitin Kumar",
year = "2018",
month = "10",
day = "9",
doi = "10.3390/risks6040113",
language = "English",
volume = "6",
journal = "Risks",
issn = "2227-9091",
publisher = "MDPI",
number = "4",

}

Business Distress Prediction Using Bayesian Logistic Model for Indian Firms. / Shrivastava, Arvind; Kumar, Kuldeep; Kumar, Nitin.

In: Risks, Vol. 6, No. 4, 113, 09.10.2018.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Business Distress Prediction Using Bayesian Logistic Model for Indian Firms

AU - Shrivastava, Arvind

AU - Kumar, Kuldeep

AU - Kumar, Nitin

PY - 2018/10/9

Y1 - 2018/10/9

N2 - The objective of the study is to perform corporate distress prediction for an emerging economy, such as India, where bankruptcy details of firms are not available. Exhaustive panel dataset extracted from Capital IQ has been employed for the purpose. Foremost, the study contributes by devising novel framework to capture incipient signs of distress for Indian firms by employing a combination of firm specific parameters. The strategy not only enables enlarging the sample of distressed firms but also enables to obtain robust results. The analysis applies both standard Logistic and Bayesian modeling to predict distressed firms in Indian corporate sector. Thereby, a comparison of predictive ability of the two approaches has been carried out. Both in-sample and out of sample evaluation reveal a consistently better predictive capability employing Bayesian methodology. The study provides useful structure to indicate the early signals of failure in Indian corporate sector that is otherwise limited in literature.

AB - The objective of the study is to perform corporate distress prediction for an emerging economy, such as India, where bankruptcy details of firms are not available. Exhaustive panel dataset extracted from Capital IQ has been employed for the purpose. Foremost, the study contributes by devising novel framework to capture incipient signs of distress for Indian firms by employing a combination of firm specific parameters. The strategy not only enables enlarging the sample of distressed firms but also enables to obtain robust results. The analysis applies both standard Logistic and Bayesian modeling to predict distressed firms in Indian corporate sector. Thereby, a comparison of predictive ability of the two approaches has been carried out. Both in-sample and out of sample evaluation reveal a consistently better predictive capability employing Bayesian methodology. The study provides useful structure to indicate the early signals of failure in Indian corporate sector that is otherwise limited in literature.

UR - http://www.scopus.com/inward/record.url?scp=85057824294&partnerID=8YFLogxK

U2 - 10.3390/risks6040113

DO - 10.3390/risks6040113

M3 - Article

VL - 6

JO - Risks

JF - Risks

SN - 2227-9091

IS - 4

M1 - 113

ER -