Machine Learning for Financial Risk Management with Python

Chapter 5. Modeling Market Risk

A measure of risk driven by historical data assumes the future will follow the pattern of the past. You need to understand the limitations of that assumption. More importantly, you need to model scenarios in which that pattern breaks down.

Miles Kennedy

Risk is ubiquitous in finance, but it is hard to quantify. First and foremost, it’s important to know how to differentiate the sources of financial risks on the grounds that it might not be a wise move to use the same tools against risks arising from different sources.

Thus, treating the various sources of financial risk differently is crucial because the impacts of those different risks, as well as the tools used to mitigate them, are completely different. Assuming that firms are subject to large market fluctuations, then all assets in their portfolios are susceptible to risk originating from these fluctuations. However, a different tool should be developed to cope with a risk emanating from customer profiles. In addition, keep in mind that different risk factors contribute significantly to asset prices. All of these examples imply that treating risk factors needs careful consideration in finance.

As was briefly discussed previously, these risks are mainly market, credit, liquidity, and operational risks. It is evident that some other types can be added to this list, but they can be thought of as subbranches of these main four risk types, which will be our focus throughout this chapter.

Market ...