How Factor Loadings Shape Credit Default Risk Estimates

# What does it mean to modify the factor loadings of a credit risk model?


# What does it mean to modify the factor loadings of a credit risk model?












I came across an example where a well-known weakness of a credit risk model was dealt with by augmenting some of the existing risk factors via increased factor loadings. This made the the model more conservative in terms of the required economic capital it recommended.

What does that mean? Is it a matter of increasing the estimated coefficients by some arbitrary percentage or is there more to it?

## Answer by phlsmk (score 5, accepted)

https://quant.stackexchange.com/a/417

"Factor loading" is a somewhat ambiguous phrase -- it could refer to the factors in a linear model (e.g. the beta in CAPM or extended linear stock models), the factors of principal component analysis, etc. If you could provide a reference to the exact example/paper it would be clearer.

In credit, however, a likely interpretation is the loadings of different macroeconomic variables and firm-specific risk components within a Cox Framework model. The Cox Framework is a generalized stochastic analysis framework where the default trigger level of a particular credit/firm is uniformly distributed, and the countdown process is controlled by a strictly decreasing firm specific function such that the firm defaults once this countdown process falls below the trigger level. Increasing the "loadings" of the risk factors within the countdown process would increase the speed at which the countdown process decreases and therefore cause the firm to default more quickly/frequently in the model. This would be one way of making a Cox-based credit model more conservative. (How much more conservative, and how to build the firm-specific countdown process realistically in the first place can be more art than exact science). For more information, the Cox framework is treated in some detail in several credit textbooks.

## Answer by Gerard Torrent (score 2)

https://quant.stackexchange.com/a/10457

It is not as simple as changing a value. You need to replace the current factor loadings by feasible values. Furthermore, factor loadings have dependencies between them, that means that when you change one of them, the other factors are affected by this change.

In the CCruncher Technical Document there is a proposal to do so. It propose to estimate the factor loadings uncertainty, and then transfer this uncertainty to the portfolio credit risk measure (EL, VaR, ES).

## Answer by quant_dev (score 1)

https://quant.stackexchange.com/a/437

In many popular copula models a factor driving a certain even (e.g. default) has the form

$ Y_i = \sum_k a_{ik} X_k + b_i Z_i $

where $\lbrace X_k \rbrace, Z_i$ are independent random factors. Coefficients $a_{ik}$ and $b_i$ are commonly called "factor loadings".

## Answer by Tom Au (score 1)

https://quant.stackexchange.com/a/11552

The "factor loadings" are really the weights attributed to different variables that predict default. If you increase the value of these factor loadings, you increase the prediction of default, thereby making the model more conservative.

Whether factor loadings are high enough ex ante is often defined by ex post events. If you had a sample of firms a certain percentage, x, of which defaulted, you might begin by adding up the predicted default rate of the sample and comparing it to x. If the predicted rate were too low/high, you would increase/decrease the factor loadings in order to get the predicted default rate to approximate the actual default rate, x.

This is an iterative process that requires a bunch of trial and error, and basically assumes that the future samples of companies will be much like the ones on which you did your adjustments of factor loadings.