General Information & Client Service
  • Americas: +1.212.553.1653
  • Asia: +852.3551.3077
  • China: +86.10.6319.6580
  • EMEA: +44.20.7772.5454
  • Japan: +81.3.5408.4100
Media Relations
  • New York: +1.212.553.0376
  • London: +44.20.7772.5456
  • Hong Kong: +852.3758.1350
  • Tokyo: +813.5408.4110
  • Sydney: +61.2.9270.8141
  • Mexico City: +001.888.779.5833
  • Buenos Aires: +0800.666.3506
  • São Paulo: +0800.891.2518

In this article, we compare the results of estimating retail portfolio risk parameters (e.g., PDs, EADs, LGDs) and simulating portfolio default losses using traditional – frequentist – methods versus Bayesian techniques. The statistical properties of the simulated risk parameter will have a significant effect on the shape of the portfolio loss distribution. Our results suggest that Bayesian estimations produce more robust estimators and result in risk parameters and loss distributions that are less volatile. Bayesian estimation has another key advantage: Posterior distributions for the model parameters can be leveraged to perform comprehensive portfolio loss simulation exercises taking into account model risk.

Default rate model set-up

We consider two common examples of retail portfolios, an auto loan book, and a credit card portfolio. Performance data is collected at a vintage/cohort level with quarterly frequency (i.e., quarterly cohorts of loans/accounts observed on a quarterly basis). The target variable to model is the vintage-level default rate, defined as the ratio of the number of accounts that have defaulted to outstanding accounts. Our methodology is in line with Licari and Suarez-Lledo (2013).1 Our target variable, (logit of) default rate, gets decomposed into three dimensions:

  1. Lifecycle component (seasoning of the accounts)
  2. Vintage quality (rank-ordering of the cohorts)
  3. Sensitivity to macroeconomic drivers

Parameters in (ii) and (iii) are assumed to be stochastic in the frequentist and Bayesian settings, while parameters (i) are assumed to be deterministic in both and fixed to the values found after performing the frequentist regression.

Model estimation results

Both frequentist and Bayesian methods produce similar average values for the key parameters. For auto loans, the results from both approaches are very similar. For credit cards, the Bayesian estimation has a significantly lower standard deviation in the parameter estimation than the frequentist standard errors, resulting in parameters that are less volatile and more precise and thus presenting a lower model risk when used in portfolio loss estimations.

Bayesian methods have the added advantage of producing posterior distributions for all parameters. The figures below illustrate the statistical shape of the posterior distribution for macroeconomic drivers and how these compare with frequentist parameters and 95% confidence intervals.

Model simulation results

The estimated models for the risk parameters are then used for the estimation of the portfolio loss distribution through a dynamic Monte Carlo simulation. Three distinct steps are considered:

  1. Macroeconomic scenarios are built using a Dynamic Stochastic General Equilibrium Model to produce forward-looking “states of the world” with quarterly updates (Licari and Ordonez-Sanz (2015)2).
  2. The Bayesian and frequentist models are then used to estimate risk parameters in each of these macroeconomic scenarios (e.g., conditional default rates).
  3. The portfolio is simulated dynamically (multiperiod default simulation) to estimate the cumulative loss distribution over nine future quarters.

For this last step, two different comparisons between the frequentist and Bayesian approaches were performed with and without model risk. The table below highlights the set-up of these two exercises, as well as key similarities and differences.

Figure 2 illustrates the set of macroeconomic scenarios used in the first step of the process for two different factors: unemployment rate and home price dynamics.

The estimated forward-looking default rates estimated in the second step of the process are shown below for both the frequentist and Bayesian approaches for the set of macroeconomic scenarios.

The first simulation exercise (macroeconomic risk only) shows consistency across both estimation methods. The simulated default rate distributions are fairly similar for both portfolios. They generate a tailed, asymmetric density with higher values at the block of stressed scenarios. The Bayesian method for credit cards seems to produce slightly less volatile default rate projections, but the overall shape of both densities is quite similar. The key difference appears when we move from exercise 1 to 2, adding model risk dimensions such as parameter volatility and error/residual properties.

Figure 4.2 drives home the key message of this section. Once we add the uncertainty coming from model risk to the simulation mechanism, frequentist and Bayesian methods produce very different outcomes. The higher precision of the Bayesian estimators flow into more concentrated, less volatile simulated default rates while still presenting the “fat-tails” that would be expected from the impact of very severe macroeconomic scenarios.

Effects on credit portfolio losses

The statistical properties of conditional default rates (conditional on a given macroeconomic scenario) influence the shape of the portfolio loss distribution. To quantify this effect, in step 3, dynamic Monte Carlo simulations are performed on both sets of conditional default rate distributions estimated in exercises 1 and 2. The charts below highlight the significant effect that these estimation results can have on the shape of portfolio losses. The severity of CCAR Adverse, Severely Adverse, ECCA S3 and S4 scenario as well as the VaR loss at 99.9% confidence levels are also shown for comparison purposes. In summary, Bayesian methods prove to be more stable, particularly after including model risk in the loss simulations.

Table 1. Macroeconomic parameters – Bayesian vs. frequentist estimations
Table 1A. Auto loan portfolio
Table 1B. Credit card portfolio
Macroeconomic parameters – Bayesian vs. frequentist estimations
Source: Moody's Analytics
Table 2. Alternative simulation exercises – macro only vs. fully-fledged
Alternative simulation exercises
Source: Moody's Analytics
Figure 1. Auto loan portfolio – posterior Bayesian distributions for macroeconomic parameters
Density functions (top) and box-plots (bottom). Red dots for frequentist betas, red vertical lines for frequentist 95% confidence intervals
Auto loan portfolio – posterior Bayesian distributions for macroeconomic parameters
Source: Moody's Analytics
Figure 2. Macroeconomic simulations – nine out-of-sample quarters
Macroeconomic simulations – nine out-of-sample quarters
Source: Moody's Analytics
Figure 3. Simulation exercise 1 – macroeconomic risk only – frequentist vs. Bayesian
Figure 3.1. Auto loan portfolio – across blocks of scenarios, five quarters out-of-sample (+Q5)
Simulation exercise 1 – macroeconomic risk only – frequentist vs. Bayesian
Figure 3.2. Credit card portfolio – across blocks of scenarios, five quarters out-of-sample (+Q5)
Credit card portfolio – across blocks of scenarios
Source: Moody's Analytics
Figure 4. Simulation exercise 2 – macroeconomic and model risks – frequentist vs. Bayesian
Figure 4.1. Auto loan portfolio – selected vintages, five quarters out-of-sample (+Q5)
macroeconomic and model risks
Figure 4.2. Credit card portfolio – selected vintages, five quarters out-of-sample (+Q5)
Credit card portfolio – selected vintages
Source: Moody's Analytics
Figure 5. Cumulative portfolio default losses – simulation exercise 1 – macroeconomic risk only – frequentist vs. Bayesian
Figure 5.1. Auto loan portfolio – after nine quarters (+Q9)
Cumulative portfolio default losses
Figure 5.2. Credit card portfolio – after nine quarters (+Q9)
Credit card portfolio – after nine quarters
Source: Moody's Analytics
Figure 6. Cumulative portfolio default losses – simulation exercise 2 – macroeconomic and model risks – frequentist vs. Bayesian
Figure 6.1. Auto loan portfolio – after nine quarters (+Q9)
Cumulative portfolio default losses
Figure 6.2. Credit card portfolio – after nine quarters (+Q9)
Credit card portfolio
Source: Moody's Analytics
Sources

1 See Licari & Suarez-Lledo, Stress Testing of Retail Credit Portfolios, Risk Perspectives Magazine, September 2013, Moody’s Analytics.

2 See Licari & Ordonez-Sanz, Multi-Period Stochastic Scenario Generation, Risk Perspectives Magazine, June 2015, Moody’s Analytics.

SUBJECT MATTER EXPERTS
As Published In:
Related Insights
Article

Dynamic Model-Building: A Proposed Variable Selection Algorithm

In this article, we propose an innovative algorithm that is well suited to building dynamic models for credit and market risk metrics, consistent with regulatory requirements around stress testing, forecasting, and IFRS 9.

Whitepaper

U.K. Residential Mortgages Risk Weights: PRA Consultation Paper CP29/16

This paper presents best practices for addressing PRA Consultation Paper CP29/16.

October 2016 Pdf Dr. Juan M. LicariDr. Dimitrios Papanastasiou, Maria Valle del Olmo
Article

Probability-Weighted Outcomes Under IFRS 9: A Macroeconomic Approach

In this article, we discuss development of a framework that addresses the forward-looking and probability-weighted aspects of IFRS 9 impairment calculation using macroeconomic forecasts. In it, we address questions around the practical use of alternative scenarios and their probabilities.

June 2016 WebPage Barnaby Black, Glenn LevineDr. Juan M. Licari
Article

Complying with IFRS 9 Impairment Calculations for Retail Portfolios

This article discusses how to address the specific challenges that IFRS 9 poses for retail portfolios, including incorporating forward-looking information into impairment models, recognizing significant increases in credit risks, and determining the length of an instrument's lifetime.

June 2016 WebPage Barnaby Black, Dr. Shirish ChinchalkarDr. Juan M. Licari
Presentation

Multi-Period Credit Risk Analysis: A Macro-Scenario Approach Presentation Slides

In this presentation, Dr. Juan Licari of Moody's Analytics will present an innovative framework for stochastic scenario generation that allows risk managers and economists to build multi-period environments, integrating conditional credit and market risk modeling to meet dynamic stress testing needs.

December 2015 Pdf Dr. Juan M. Licari
Presentation

Market Risk Stress Testing Models Presentation Slides

In this presentation, Dr. Juan Licari presents a two-stage process that generates consistent, transparent scenario-specific forecasts for all relevant market and credit risk instruments, ensuring cross-consistency between projections for macroeconomic and financial series.

December 2015 Pdf Dr. Juan M. Licari
Webinar-on-Demand

Market Risk Stress Testing Models

In this presentation we present a two-stage process that generates consistent, transparent scenario-specific forecasts for all relevant market and credit risk instruments, ensuring cross-consistency between projections for macroeconomic and financial series.

December 2015 WebPage Dr. Juan M. Licari
Webinar-on-Demand

Multi-Period Credit Risk Analysis: A Macro-Scenario Approach

In this presentation, we present an innovative framework for stochastic scenario generation that allows risk managers and economists to build multi-period environments, integrating conditional credit and market risk modeling to meet dynamic stress testing needs.

November 2015 WebPage Dr. Juan M. Licari
Webinar-on-Demand

IFRS 9 Impairment Webinar Series – Models for Implementation

This webinar discusses determining the best approaches for model development and governance for IFRS 9 Impairment calculations.

September 2015 WebPage Manuele Iorio, Dr. Juan M. Licari
RESULTS 1 - 10 OF 24