What is Value at Risk? VaR and Risk Management
Vložit
- čas přidán 17. 06. 2019
- In todays video we learn about Value at Risk (VaR) and how is it calculated?
Buy The Book Here: amzn.to/37HIdEB
Follow Patrick on Twitter Here: / patrickeboyle
What Is Value at Risk (VaR)?
Value at risk (VaR) is a calculation that aims to quantify the level of financial risk within a firm, portfolio or position over a specific time frame. This metric is most commonly used by investment and commercial banks to determine the extent and occurrence ratio of potential losses in their institutional portfolios.
Risk managers use VaR to measure and control the level of risk exposure. One can apply VaR calculations to specific positions or whole portfolios or to measure firm-wide risk exposure.
VaR modeling aims to calculate the potential for loss in the portfolio being assessed and the probability of occurrence for the defined loss. One measures VaR by assessing the amount of potential loss, the probability of occurrence for the amount of loss, and the timeframe involved.
A VaR calculation based on data from a period of low volatility may understate the potential for risk events to occur and the magnitude of those events. Risk may be further understated using normal distribution probabilities, which rarely account for extreme or black-swan events.
The financial crisis of 2008 exposed many of the problems with VaR as relatively benign VaR calculations understated the potential occurrence of loss events posed by portfolios of subprime mortgages. Risk was underestimated, which resulted in extreme leverage ratios within subprime portfolios. As a result, the underestimations of occurrence and risk magnitude left institutions unable to cover billions of dollars in losses as subprime mortgage values collapsed.
Risk Management
Patrick, I am currently preparing for an interview for a market risk analyst opportunity and this is the most helpful content ever. I truly appreciate your help and will def check out your book when I get my contract and first paycheck
so underrated );
Hi Patrick
I really like your current video content and your older academic content. Would you ever consider doing a second channel for new academic videos? They've actually helped me more that a few times at work. Currently watching this one for an intro for a FX VAR project
Can't believe this doesn't have more views
Great video Mr. Boyle as usual, currently reading through your "stats.. for Traders" up to CH11, really enjoying it too, quick question, are the answers to the questions at the back of the chapters available somewhere, I would be particularly interested in the answers for the calculation based questions.
Unfortunately they are not, but I might do a solutions book at some point in the future.
Thanks for the quick response Mr. Boyle, maybe a video solution for the various chapters might be easier to coordinate, could be incrementally released as well whenever time permits. I think majority of your followers (myself included) would appreciate that.
Great video, Hi from Canada... I have a book called "A Wealth of Common Sense" by Ben Carlson. What's your view of this book Patrick? Is it a good read for starters?
I have not read the book, but I have heard good things about Ben. I'll have to add it to my reading list.
Hi Patrick, thanks for covering the topic of VaR, I appreciate your discussion on the topic!
However, I believe there are some issues with the way you presented the statistical interpretation of VaR. Assuming a normal distribution (as presented in the charts in the video) and a confidence (CI) interval of 95%, the statical probabilities are actually 2.5% - not 5%. This is due to the two tails of the normal distribution, and the 5% being an expression of extremes (>2 Standard Deviations, both positive and negative). Since there are two extreme outcomes, we get (100%-95%)/2 = 2.5% probability of an extreme positive outcome and a 2.5% probability of an extreme negative outcome. To frame this in the context of your video:
With a CI of 95%, we expect that the most we can lose in a day is $1 M, so long as we are willing to accept that 2.5% of the time our losses will be greater than $1 M on a given day.
The same probabilities also hold for outperformance on the positive end, however VaR is only concerned with loss. I would agree it is correct to say that we anticipate to see extreme loss/return only 1/0.05 of the time, or once in every 20 days in the cycle. More precisely, we expect to see an outsized positive/negative return 1/0.025 of the time, or once every 40 days in the cycle. However as you stated the reality of what may actually happen is also very different from what the stats tell us to anticipate 😊. This is also despite other VaR issues such as non-normal return distribution, which has often proven to be the case in the markets.
Hope you found this comment useful, and thanks for all the great content!
^This.
you can perform a standard coverage test to see at what confidence level your vcv VaR fits the return distributions