Hi everyone! Important Clarification: For a detailed explanation of how the Darwinex Risk Engine calculates and uses VaR, kindly watch the following 3 videos: 1) Calculating the Risk (D-Leverage) of a Position: ua-cam.com/video/QJMBXabXMLw/v-deo.html 2) How Value-at-Risk (VaR) is calculated at Darwinex - Part 1: ua-cam.com/video/rgxXjdEnJq8/v-deo.html 3) How Value-at-Risk (VaR) is calculated at Darwinex - Part 2: ua-cam.com/video/_IsLJbn2B1o/v-deo.html Many thanks! All the best, The Darwinex Team
Hi Martin, thank you for all your videos about VaR and because of that, i went to read more about VaR is about. I am a Masters students in Economics, therefore statistics is really interesting to me. I want to point out, a very important part of the VaR calculation is the standard deviation of returns. However, you are using the price change ratio in the series as the standard deviation of returns. For your case, I think you assume that positions will only take on a value of gain/loss based on the price movement of the assets. However in reality, most traders do have TP and SL in place, therefore the probability density function of the returns should not be assumed to be normal. For example, if the SL is set at a very tight to the entry price, then a huge majority of losses (maybe >90%) will be at the amount risked calculated based on the lot size and stop loss values. Same goes for profit, if there is no TP set, perhaps returns will approach a normal distribution, but if you set a TP, there will be a significant portion of returns at the TP value, hence your distribution will be nothing close to normal. In this case, if the only entry/exit rule will be based on TP and SL, a binomial distribution would be more accurate to describe the returns. However, other weird distributions will arise when you have different exit rules, such as using a trailing stop for winners and fixing a stop loss for losers, or perhaps just put a stop loss as an emergency and exit based on other settings.
I am interested to know how we can deal with value at risk with trades going from 5 minutes to 1 month (because you have to compare things comparable when compute var) ? Maybe deal with 3 portfolios short,medium, long as the impact between each others will be negligible ? Many tks for high quality content !
Hi everyone!
Important Clarification: For a detailed explanation of how the Darwinex Risk Engine calculates and uses VaR, kindly watch the following 3 videos:
1) Calculating the Risk (D-Leverage) of a Position:
ua-cam.com/video/QJMBXabXMLw/v-deo.html
2) How Value-at-Risk (VaR) is calculated at Darwinex - Part 1:
ua-cam.com/video/rgxXjdEnJq8/v-deo.html
3) How Value-at-Risk (VaR) is calculated at Darwinex - Part 2:
ua-cam.com/video/_IsLJbn2B1o/v-deo.html
Many thanks!
All the best,
The Darwinex Team
Thank you again Martyn, .
Unbelievable information for free.
I'm glad you are benefitting. Many thanks for your feedback Alireza, on this and all the other videos. Martyn
Hi Martin, thank you for all your videos about VaR and because of that, i went to read more about VaR is about. I am a Masters students in Economics, therefore statistics is really interesting to me. I want to point out, a very important part of the VaR calculation is the standard deviation of returns. However, you are using the price change ratio in the series as the standard deviation of returns. For your case, I think you assume that positions will only take on a value of gain/loss based on the price movement of the assets. However in reality, most traders do have TP and SL in place, therefore the probability density function of the returns should not be assumed to be normal. For example, if the SL is set at a very tight to the entry price, then a huge majority of losses (maybe >90%) will be at the amount risked calculated based on the lot size and stop loss values. Same goes for profit, if there is no TP set, perhaps returns will approach a normal distribution, but if you set a TP, there will be a significant portion of returns at the TP value, hence your distribution will be nothing close to normal.
In this case, if the only entry/exit rule will be based on TP and SL, a binomial distribution would be more accurate to describe the returns. However, other weird distributions will arise when you have different exit rules, such as using a trailing stop for winners and fixing a stop loss for losers, or perhaps just put a stop loss as an emergency and exit based on other settings.
I came here to learn about VAR because Ramesh in the movie "Margin Call" mentioned it lol
simple illustration of a nice concept
Very helpful 👌
Thanks Fouad
I am interested to know how we can deal with value at risk with trades going from 5 minutes to 1 month (because you have to compare things comparable when compute var) ? Maybe deal with 3 portfolios short,medium, long as the impact between each others will be negligible ? Many tks for high quality content !