7. Other Methods to Estimate the Probability of Default
7. Other Methods to Estimate the Probability of Default
1. Experts-Based, Statistical-Based, and Numerical Approaches
Credit quality analysis from an experts-based approach will apply frameworks such as the four Cs of credit (Character, Capital, Coverage, Collateral) , LAPS (Liquidity, Activity, Proitability, Structure) ,and CAMELS (Capital Adequacy, Asset Quality, Management, Earnings, Liquidity, Sensitivity) . Qualitative features need to be factored into any analysis along with quantitative components.
Statistical-based classification centers on the fact that a quantitative model is essentially just a description of the real world within a controlled environment. Models are simply used to express a viewpoint of how the world will likely behave given certain criteria. A quantitative model will have a qualitative (formal) formulation that describes the basic view of the world we are trying to capture in the model; it will also have the underlying assumptions needed to build the model. The assumptions, which serve to simplify the process, should cover organizational behavior, possible economic events, and predictions on how market participants will react to these events.
Numerical approaches have the objective of deriving optimal solutions using “trained” algorithms and incorporating decisions based on relatively weak information in very complex environments. An example is a “neural network,” which is able to continuously update itself for changes to the environment.
2. Agencies’ Ratings vs. Experts-Based Approaches
A rating agency’s assignment processes will be different than the internal classification methods used by banks, even though the underlying processes are often analogous.
Relative to a formal approach, such as quantitative analysis based on statistical models, experts-based approaches are neither considered to be inferior nor superior. An experts-based approach relying on judgment will require significant experience and repetitions in order for many judgments to converge. Also, the challenges of such an approach include the dynamic nature of organizational patterns; M&A activity, which blends portfolios and processes; and changing company cultures.
A predictive performance that may work in one period is not necessarily indicative of future performance. Also, internal credit rating systems are difficult and time-consuming to develop. However, having a reliable internal system represents a significant value added for an entity.
3. Statistical-based Approaches
3.1 Structural and Reduced Form Approaches
The foundation of a structural approach (eg, the Merton model) is the financial and economic theoretical assumptions that describe the overall path to default. Under this approach, building a model involves estimating the formal relationships that link the relevant variables of the model.
Reduced form models (e.g, statistical and numerical approaches) arrive at a final solution using the set of variables that is most statistically suitable without factoring in the theoretical or conceptual causal relationships among variables.
- A reduced form model will not make any ex ante assumptions about causal drivers for default (unlike structural models)
- Specific firm characteristics are linked to default, using statistics to tie them to default data.
- The independent variables in these models are combined based on their estimated contribution to the final result and can change in terms of relevance depending on firm size, firm sector, and economic cycle stage.
- A significant model risk in reduced form approaches results from a model’s dependency on the sample used to estimate it. To derive valid results, there must be a strong level of homogeneity between the sample and the population to which the model is applied.
Reduced form models used for credit risk can be classified into statistical and numerical-based categories.
- Statistical-based models use variables and relations that are selected and calibrated by statistical procedures.
- Numerical-based approaches use algorithms that connect actual defaults with observed variables.
- Both approaches can aggregate profiles, such as industry, sector, size, location, capitalization, and form of incorporation, into homogeneous “top-down” segment classifications.
- A “bottom-up” approach may also be used, which would classify variables based on case-by-case impacts. While numerical and statistical methods are primarily considered bottom-up approaches, experts-based approaches tend to be the most bottom up.