A Survival Approach to Prediction of Default Drivers for Indian Listed Companies

Abstract

The objective of the research study is to identify the key predictors that can explain default risk for Indian listed companies using survival analysis. The author has applied the semi-parametric Cox proportional hazard model to test the impact of financial ratios, capital market ratios, macro-economic variables, size and age of companies, and the ownership structure of promoters to a dataset of 859 companies panning across 10 sectors. Unlike traditional models on default prediction, survival models focus on "time to default" as the dependent variable. The empirical findings reveal that return on capital employed (ROCE), return on net worth (ROE), interest coverage ratio, exchange rate volatility, GDP growth rate, stock index, promoters holdings % and the percent of shares pledged are all significant predictors of default. Among the market variables, it is seen that beta and the ratio of market value of equity/book value of debt are statistically significant variables in explaining default risk. The empirical findings also generate the hazard ratio for each covariate which examines the predicted change in the hazard for a unit increase in the predictor. The author extends the research by applying the marketbased KMV structural model to obtain continuous observations of default probability and regressing the same against all the 1 covariates (Gupta et al.,. It is observed that the set of significant covariates are almost common to those generated by our survival approach. The study concludes in emphasizing the significance of survival models in default prediction as unlike traditional accounting-based and market-based models, these models assess relationship between survival time and covariates. The application of survival models is strongly recommended for credit risk evaluation and modeling as structuring of loans can be done by lenders by assessing the survival times of different firms across the entire observation period being considered.

Modeling corporate financial distress using financial and non-financial variables

“…A substantial number of Indian companies are owned by family houses and corporate groups, who are the promoters. High promoters stake or ownership concentration reduce the risk of companies going into financial distress, increase firm value and decrease agency costs (Gupta, 2017; Kumar and Singh, 2013; Ciampi, 2015; Manzaneque et al , 2016b).…”

“…High stock pledge ratio from promoters of the firm can prove to be early warning signs of financial distress as it creates personal leverage. The increase in the degree of personal leverage and overinvestment in the stock market by the promoters also create risk for the companies to a certain degree, which in turn weakens the corporate governance structure (Gupta, 2017; Lee and Yeh, 2004).…”

“…Gupta (2017) used a data set of 859 Indian listed companies across ten sectors and applied a cox proportional-hazard model to identify the key drivers that can explain the default risk. The author found that return on capital employed, return on net worth, interest coverage ratio, exchange rate volatility, GDP growth rate, stock index, the percentage of promoter holdings and the per cent of shares pledged are all significant predictors of default.…”

Prediction Model as Sustainability Tool for Assessing Financial Status of Non-Profit Organizations in the Slovak Republic

“…Finally, we verify the proposed model on the testing sample. Gupta (2017) [32] argues that the validity of a model is assessed using ROC. If the metric is equal to 1, they indicate perfect (ideal) predictive accuracy.…”

Comparison of Survival Analysis Approaches to Modelling Credit Risks

“…The ratio of core capital to total risk-weighted assets also increased slightly from 16.05 percent in December 2017 to 16.15 percent as at March 2018 [11]. Historically, lending institutions have used different models in managing credit risk [12][13][14][15][16][17]. Several authors have looked at credit risk scoring and modelling by extending CPH models [2,18,[20][21][22][23][24].…”