| Credit Metrics is a method of reigning in credit risk by | | | | Although credit metrics and risk metrics are similar in |
| modeling changes in credit ratings portfolio. This | | | | many ways they are not the same. The primary |
| implies a propositional change in value of the holdings. | | | | difference between the two is that risk metrics |
| Credit metrics tries to construct that is not readily | | | | presents an loads of daily liquid pricing data which can |
| observable, which is the volatility of value due to | | | | be easily used to construct a model of conditional |
| changing credit quality. This approach renders credit | | | | volatility. On the other hand credit metrics offers |
| metrics more of an exercise in proposing models and | | | | relatively less and sporadically priced data for |
| which explain the changes in credit related | | | | constructing a model of unconditional volatility |
| instruments. More than often the models that best | | | | The recovery of a claim remains unknown until an |
| describe credit risk don't rely on the assumption that | | | | obligor defaults. Credit metrics on the other hand |
| returns distribution is imperative. | | | | models recovery by using a beta distribution. A beta |
| Credit metrics is basically a framework that helps to | | | | distribution is characterized by a mean and standard |
| quantify credit risk on portfolio of everyday credit | | | | deviation. The recovery of the distribution is affected |
| products. This includes loans, commitments to lend, | | | | by changes in parameters as demonstrated by the |
| and market -driven instruments which are vulnerable | | | | beta distribution spreadsheet. |
| to counterparty defaults. The sound of knowledge of | | | | In credit metrics the changes in value is not only |
| Credit metrics enables you get a transparent | | | | influenced by chancy default events but also by the |
| depiction of credit risk. Transparency and effective | | | | upswings and downswings in credit quality. Credit risk |
| management share a direct proposition and usually | | | | also addresses the value-at-risk (VaR) which is |
| goes hand in glove. The common crisis that has been | | | | basically the volatility of value and not just the |
| plaguing the credit risk measurement is the absence | | | | expected losses. It makes sense to address the |
| of a common point reference. The multiple | | | | co-relation of credit quality fluctuation across obligors |
| approaches to measure of credit risk render them | | | | as it allows you directly calculate the potential over |
| practically incomparable. | | | | -concentration across the portfolio. |
| Credit measure and Credit metrics are often | | | | Modeling transitions for a single name is pretty simple. |
| misinterpreted to be the same. When we refer to a | | | | If one has an idea of the probability to each state, |
| measure we are actually assigning a number to | | | | then he/she can approximately simulate a transition |
| something. A metric on the other hand is how | | | | corresponding to each state by observing a random |
| interpret that assigned number. A simple example | | | | uniform variable. The transition can be made by |
| would be that of calculating a person's height. Let's | | | | basing on the outcome of the random uniform |
| ay it measures to 5.1 inches, the inches is the | | | | variable. The glitch is when there are multiple |
| measure of the person's height and the, "height" is | | | | correlated names in the portfolio. |
| the metric. | | | | |