Backtesting Value-at-Risk: Standard coverage test (Excel)

How one can evaluate whether a particular Value-at-Risk model is appropriate? Today we address this question and investigate the simplest backtesting procedure for VaR - the standard coverage test.
Don't forget to subscribe to NEDL and give this video a thumbs up for more videos in Econometrics!
Please consider supporting NEDL on Patreon: / nedleducation

[@NEDLeducation]

You can find the spreadsheets for this video and some additional materials here: drive.google.com/drive/folders/1sP40IW0p0w5IETCgo464uhDFfdyR6rh7
Please consider supporting NEDL on Patreon: www.patreon.com/NEDLeducation

[@kessler88]

Excellent video mate - thanks!

[@finalpurez]

Thank you once again!

[@MikFrey]

Well paced video with good example :)
Other videos about var explain less in 30 minutes :)

[@dorisallamani4199]

Thank you for your video! Why is the return of the stock calculated as the Product of 1+daily returns in total squared to 1/sample size as apposed to beeing calculated as the mean of daily returns?

[@JS_quinonez]

Excellent work! This serie has been helpful for some task at work. Do you have any documentation about the test?

[@robertmota1851]

Hi NELD, thanks for the videos. but i didn't understand z test for alfa level confidence why this formulas works or why is that.
Thanks for every think again.

[@LorenzoMastromattei]

Good! One question: the variance-covariance VaR (-2.45%) is a daily VaR, for example is 05/02/2023 VaR? considering the backtest we test entirely the last n days losses over the daily VaR ? I'm confused, thanks!

[@drek273]

Hello NEDL. How do we conduct standad coverage tests on other VaR models? can you show us how that is done?

[@ghulamnabi6331]

hi. Can you help me how to calculate the monthly standard deviation of the stock market from daily returns

[@NEDLeducation]

Hi Ghulam, and thanks for the question! Generally, you can calculate monthly volatility from daily standard deviation by simply multiplying it by the square root of a number of trading days in a month (most commonly 21). So just multiply daily volatility by SQRT(21).

[@NEDLeducation]

This assumes return independence, however. For more advanced (and rarely used, tbh) applications see this: ruclips.net/video/_z-08wZUfBc/видео.html

[@vincenzocardone510]

Hi can you advise a good tutorial for VaR backtesting with traffic lights and how to replace historical returns with actual ones ? Many thanks

[@NEDLeducation]

Hi Vincenzo, and thanks for the question! The traffic lights system is actually implemented in the standard coverage test. You could say the colour is green if p-value is above 10%, yellow if it is between 1% and 10%, and red if it is below 1%. For backtesting with "real returns", you would just need to substitute the parameters to equal those estimated in-sample and test for violations on out-of-sample returns.

[@vincenzocardone510]

@@NEDLeducation hi and if I want ti calculate VaR with historical simulation, how to use in the standard coverage test?

[@NEDLeducation]

@@vincenzocardone510 Hi Vincenzo, and thanks for the follow-up question! The video on that is along the way :)

[@lklkos8928]

master

[@Hemzees]

Hi could you do a video on calculating climate value at risk? This is becoming increasingly important in my field and I can't find any valid method on youtube yet

[@mohammedmonaf6128]

Thnx for your great work.. Realy help.. Actually I have a question in risk management, can I send you email to explain the question?