Welcome to our Week in Risk blog! Stop by our forum to join in on the FRM discussions, and visit our YouTube channel to view in-depth videos that David Harper posts weekly, including his newest R programming playlist! This week, we’ve included our newest FRM practice questions, forum discussions, and other great risk articles that we hope you will find interesting. Have a great week!
New Practice Questions
- Valuation & Risk Models: P1.T4.910. Barbells and bullets (Tuckman Ch.4) https://trtl.bz/2TBvHoc Bond duration and convexity are fundamental fixed income (FI) concepts, but most people–even in finance–struggle to understand them. To master them, practice their calculation! Here we’ve added fresh practice questions to our extensive FI database; practice here will make you sharp on the difference between modified, Macaulay and effective duration/convexity measures.
- Credit Risk Measurement & Management: P2.T6.901. Credit exposure and valuation adjustments (Gregory, Ch.4) https://trtl.bz/2u6HSKp Here are three new questions in the counterparty fundamentals sequence. What we’re doing here (based on Gregory’s Chapter 4) is reinforcing the foundation. Gregory gets difficult in subsequent chapters, but his complexity is a build on the fundamental ingredients; e.g., current versus credit exposure, replacement cost, potential future exposure (PFE), credit value adjustment (CVA). Did you know that Jon Gregory himself registered and posted in the forum last week (because we found a typo for him)? We’re proud to count him as a member.
- Valuation & Risk Models: Market maker’s delta-hedge illustrated (T4-20) https://trtl.bz/2TJ0ejH This video concludes a sequence (the playlist includes twenty videos per the “T4-20” tag) on option Greeks and delta-gamma hedging. This video illustrates–visually and numerically–exactly how a market market who is short position gamma is exposed to volatility.
In the Forum (highlights only)
- Key math: This week saw some smart mathematical questions asked in the forum (although no question is ever dumb, I mean that sincerely). You may know that the FRM has a quantitative hurdle? I am proud of that, if for no other reason that quantitative skills are only becoming more important. Most of us will never be financial engineers, but many of us will wear the “citizen data scientist” hat at various points in our career. Here are three examples of key math in the past week.
- First, Catherine asked about the two different use cases of gamma (one as a second derivative, another as the first derivative of delta): https://trtl.bz/2Jj6aLr. One appearance utilizes the Taylor Series approximation, which is a formula that unifies several asset classes in the FRM (bonds, options, portfolios, and even CVA in Gregory’s counterparty risk). The other use case concerns delta (and gamma) neutralization, a difficult idea that I recently covered in a video here https://trtl.bz/2FjvA7K (as part of a playlist, there are actually several videos in the playlist that include coverage of option gamma)
- Second, Merlinius made a good point about the definition of a positive definite matrix here https://trtl.bz/2JnbUnx (it links to an XLS where I show a violation); the nice thing about understanding this is that you are sort of forced to learn how portfolio variance is calculated.
- Third, thank you, James, for posting some calculator questions here https://trtl.bz/2TOKUS5. I do freely admit that we have so far under-served our audience in regard to calculator usage. But it’s important to be proficient with the calculator, if for no other reason than to save time on the exam. This made me realize that I want to post some calculator usage videos, so those are coming soon in the next playlist.
- Variance-bias tradeoff: Frequent forum contributor, gprisby, asked about the variance-bias tradeoff here https://trtl.bz/2TQzsFA. Before you think this concept dull, can I tell you that in January I attended a data science (machine learning) bootcamp in Austin, Texas. The syllabus included the variance-bias trade-off, which turns out to be an important issue in machine prediction. I was surprisingly well-prepared for it due to my exposure to extreme value theory (EVT) in the syllabus. The syllabus assignment is Kevin Dowd book, Measuring Market Risk. I think this book has been in the syllabus since I started teaching the FRM and my appreciation for it grows every year. Unlike almost every other book in the syllabus, I don’t think any candidate has ever identified an error in this book. Amazing! His precision is almost unparalleled and this text remains one of the best introductions to market risk. It is a fantastic paring, if you will, with Carol Alexander’s 4-volume masterpiece (much of which overlaps with the PRM syllabus, not by accident).
- Default probability terminology: We had two questions about a perennial topic, default probabilities. Here’s the thing: we must be careful about the terms.
- Gerard posted a default probability (PD) question set that we didn’t write and GARP definitely didn’t write. I sometimes answer external questions that are bad because they can be instructive, as is this one at https://trtl.bz/2TZ8kUY. The terms are imprecise, so the question is not helping you build a useful habit. (There are a lot of imprecise questions floating around, a bad question can really waste your time.) The FRM employs consistent definitions for conditional PD (e.g., hazard rate), joint/unconditional PD, and cumulative PD. The forum has dozens, if not hundreds, of exercises using these terms, including their synonyms.
- The other post this week asked about an old FRM practice exam PD-based question item (https://trtl.bz/2TVFepf) and illustrates why you should not practice on old exams: this question uses imprecise language before GARP refined the PD terminology for better clarity and consistency. I’m proud to say we (our subscribers) contributed a lot of the feedback that helped input-wide to improve the syllabus glossary with respect to the PD concept cluster. In any case, please trust me and don’t practice on old exams, there are many concepts that have been refined in recent years.
- Risk management: This is one of the better articles I’ve read about the deadly Boeing 737 Max 8 crashes https://trtl.bz/2TXzbQY. This should be a case study in risk management, and it has parallels to financial risk. The Seattle Times reports, “Going against a long Boeing tradition of giving the pilot complete control of the aircraft, the MAX’s new MCAS automatic flight control system was designed to act in the background, without pilot input.” Apparently, the plane’s computer system kept seizing control away from the pilots. Is this not an issue for our times? How much control should our systems have in this new world of human-machine collaboration; e.g., high-frequency trading? Further, “The FAA, citing lack of funding and resources, has over the years delegated increasing authority to Boeing to take on more of the work of certifying the safety of its own airplanes.” This sort of self-auditing is a classic problem and is hardly the only governance mistake cited in the article. Governance isn’t sexy and doesn’t always get much attention, but most of our case studies include governance failures.
- Two of the best financial bloggers: Can I share my two favorite financial bloggers, if only because they turned in another great week? Matt Levine at Bloomberg somehow writes deeply and informatively every weekday; e.g., Monday’s discussion of narrow banks https://trtl.bz/2JkXmVC. In recent weeks, he has covered various credit default swap (CDS) situations like nobody’s business; e.g., https://trtl.bz/2TwZQnr. Felix Salmon posts Axios Edge every Sunday at https://trtl.bz/2JkzGAG. For my money, Felix is the best financial blogger in the business. He hosts the Slate Money podcast https://slate.com/podcasts/slate-money.
- My article in Seeking Alpha: Finally, my latest article was published on Seeking Alpha where I am a contributor: it’s a review of my position in the Austin-based fintech company Q2 (QTWO) Holdings (https://trtl.bz/2Jlkd3r). Can you guess my favorite city, btw? My academic training and teaching have prepared me well for buying stocks, but in the article I share that I have not developed a consistent sell discipline. I find selling much more difficult than buying. At the end of the article I wrote: “My view of fintech is influenced by a framework articulated in an excellent paper that is assigned to Financial Risk Manager (FRM) candidates. My day job is teaching the FRM syllabus. The paper is called ‘On the Fintech Revolution: Interpreting the Forces of Innovation, Disruption and Transformation in Financial Services’ by Gomber, Kauffman, Parker and Weber.” Isn’t that cool? The FRM syllabus has actually influenced my real portfolio! My QTWO is up +60% but don’t follow me here on this one currently, it’s fully valued (no current margin of safety), and I’m all about the risk management, as you’d expect. I did double-down on housing in December, and so far I’ve been lucky on that call. But that’s another story.