Multiperiod Corporate Default Prediction – A Forward Intensity Approach
Author: Jin-Chuan DUAN, Jie SUN, and Tao WANG
Methodological Foundation of DeepCredit iEWS
Published:Sep 2012. Journal of Econometrics 170(1), 191-209 (Theoretical Foundation of DeepCreditTM EWS)
This paper proposes a forward intensity approach for default prediction over different future periods. This bottom-up approach learns from big data with endogenous and exogenous factors to build a sophisticated multi-dimensional mapping from a firm’s current risk profile to its default dynamics, and thus produces a forward-looking point-in-time PD term structure with high granularity and precision.
Proxy CDS Curves for Individual Corporates Globally
Author: Jin-Chuan DUAN
Methodological Foundation of DeepCredit pCDScurve
Published: First Draft Sep 2017. This Version Mar 2020. Working Paper
This paper offers an intuitive, practical and robust predictive regression model linking liquid USD denominated CDS premiums of different tenors to a set of obligor-specific attributes. This model can generate proxy CDS curves for corporates without liquid or traded CDS.
Notably, the proposed method can resolve the lack of data in FRTB CVA calculation.
Assessing Corporate Vulnerabilities in Indonesia: A Bottom-Up Default Analysis
Authors: Jorge A. CHAN-LAU, Weimin MIAO, Ken MIYAJIMA, and Jongsoon SHIN
Published: Apr 2017. IMF Working Paper WP/17/97
This paper assesses corporate vulnerability in Indonesia by using the BuDA (Bottom-up Default Analysis) approach. The results suggest risk in the corporate sector are manageable even if the economy were hit by a protracted recession accompanied by a large currency depreciation, and the authorities have been proactive in monitoring corporate vulnerabilities and encouraging proper currency risk management.
Dynamic Point-in-Time Probabilities of Default via Stochastic Rating Migration
Published: Oct 2016. Working Paper
This paper proposes a stochastic rating migration model constructed with a set of credit cycle indices to reflect market conditions globally and by sectors, and then uses this rating migration model to deduce any forward PIT PDs that naturally become dynamic over time and reflective of a specific forward period of interest.
Notably, the proposed approach is ready for producing the lifetime PIT PD term structure for both stress testing and IFRS 9/CECL impairment.
Default Correlations and Large-Portfolio Credit Analysis
Authors:Jin-Chuan DUAN and Weimin MIAO
Published: Sep 2016. Journal of Business & Economic Statistics, 34(4), 536-546
This paper proposes a novel factor model for generating default correlations via a low-rank plus sparse representation. This model can be used to produce and stress portfolio credit risk profiles, say default-rate and portfolio-loss distributions.
Stress Testing with a Bottom-Up Corporate Default Prediction Model
Authors: Jin-Chuan DUAN, Weimin MIAO, and Tao WANG
Methodological Foundation of the BuDA (Bottom-up Default Analysis) analytical tool for IMF
Published: Aug 2014. Working Paper
This paper proposes a readily implementable bottom-up credit stress testing method that can translate adverse macroeconomic scenarios into different distributions of corporate defaults for an economy and/or any other target portfolio of interest. It also provides a novel way of estimating the stress-testing regressions with mixed-frequency data.
Author: Jin-Chuan DUAN, Baeho Kim, Woojin Kim, and Donghwa Shin
Published: Aug 2018. Journal of Banking and Finance 94, 235-250
This paper proposes a new methodology for estimating default term structure for private firms leveraging the forward intensity model. As private firms do not have traded stock prices, this paper devises a methodology to obtain a public-firm equivalent distance-to-default by projection that reference the distance-to-defaults of public firms with comparable attributes. It is the first study to investigate the dynamic behavior of default risk for private firms over different future horizons