PREDICTION OF CORPORATE DEFAULT USING LOGISTIC REGRESSION

Peer reviewed: 
No, item is not peer reviewed.
Scholarly level: 
Graduate student (Masters)
Date created: 
2019-12
Keywords: 
Default
Logistic regression
Marginal effect
Relative or receiver operating characteristic
Merton’s Distance-to-Default
Abstract: 

The main aim of the research is to examine the importance of Merton's (1974) distance-to- default measure in predicting corporate defaults. The data sample includes 75,667 companies from 1975 to 2007. We compare the predictive power of Merton's distance-to- default to accounting variables used in Ohlson (1980), Altman (1968), and a set of market measures used in Campbell et al. (2008).

The marginal effect is used to evaluate the efficiency of the independent variables to forecast corporate defaults. The relative or receiver operating characteristic (ROC) curve is used to show the accuracy of the model. The findings show that Merton distance to default improves the efficiency of the model and has a high marginal effect among the independent variables, as shown in the paper.

Description: 

MSc in Finance Project-Simon Fraser University.

Language: 
English
Document type: 
Graduating extended essay / Research project
Rights: 
Copyright remains with the author.
File(s): 
Senior supervisor: 
Dr. Deniz Anginer
Department: 
Beedie School of Business-Segal Graduate School
Statistics: