Sharpe Ratio - is it correct? - page 2
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
You are 100% right cymact :) I still can't get used to the new way the orders are handled
Anyway, with new calculation SR is 0.16 :-)
Just an update. I got a response that this bug has been fixed. Can't wait for the new version!
Just an update. I got a response that this bug has been fixed. Can't wait for the new version!
I see new build 298 has revised calculation values for Sharpe Ratio. Is this description correct from Mathematics in Trading: How to Estimate Trade Results ? It now appears to be very difficult to get a Sharpe Ratio >=3.0.
I see new build 298 has revised calculation values for Sharpe Ratio. Is this description correct from Mathematics in Trading: How to Estimate Trade Results ? It now appears to be very difficult to get a Sharpe Ratio >=3.0.
That's great. I downloaded it but my EA stopped working after I performed an update so I couldn't test my EA :-)
Professionals (who by the way use Sharpe Ratio or K-ratio and say that Profit Factor and other MT4 ranks are useless) say that with Sharpe Ratio more than 1 you are quite sure to have a reliable system that you can leverage quite much without much risk. System with SR over 2 makes you a millionaire in quite a short period of time. Sharpe Ratio shows consistancy of profits and losses. If losses are rare and very small and profits are bigger than losses and consistent then it goes up, but if your system allows a bigger loss once in a while then SR goes down dramatically. Unfortunately the same applies to profits. If you have small or no profits and you wait for a big hit once in a while, the SR will be rather small.
Unfortunately we still have our Tester in an "optimistic" mode, where spread is not applied (or is very small) so I wouldn't trust the results 100%, even with the corrected SR formula
That's great. I downloaded it but my EA stopped working after I performed an update so I couldn't test my EA :-)
Professionals (who by the way use Sharpe Ratio or K-ratio and say that Profit Factor and other MT4 ranks are useless) say that with Sharpe Ratio more than 1 you are quite sure to have a reliable system that you can leverage quite much without much risk. System with SR over 2 makes you a millionaire in quite a short period of time. Sharpe Ratio shows consistancy of profits and losses. If losses are rare and very small and profits are bigger than losses and consistent then it goes up, but if your system allows a bigger loss once in a while then SR goes down dramatically. Unfortunately the same applies to profits. If you have small or no profits and you wait for a big hit once in a while, the SR will be rather small.
Unfortunately we still have our Tester in an "optimistic" mode, where spread is not applied (or is very small) so I wouldn't trust the results 100%, even with the corrected SR formula
Thanks. Do we know if the MT-5 uses the individual deals or just the positions for calculating the Sharpe Ratio (SR). Because with Martingale type EAs, with multiple deals in a single position, there could be, for example, 1 profitable deal and 1 losing deal in the position, but position closed in profit. This would make a big difference in how SR is interpreted.
Individual deals of INOUT and OUT types
Is the formula given here (http://championship.mql4.com/2007/news/203) still used by MT5 to calculate Sharpe Ratio?
I've tried to check it by analyzing several reports with 2-3 deals and the results I'm getting with the formula differ from the Sharpe Ratio in the report. Or does it use some non-zero risk-free rate (RFR)?
risk free rate assumed to 0
sharpe_ratio=(ahpr-(1.0+0.0))/std_dev
ahpr is average holding period return. Calculated as sum of hpr divided on hpr total.
hpr calculated as ratio between balance after out or inout operation and previous balance (after_balance / prev_balance). updated after enivid's remark
std_dev is standard deviation of all hprs from the average hpr
stringo:
hpr calculated as difference between balance after out or inout operation and previous balance.
A difference or a ratio? For example, if the balance was $10,000 and became $11,000, HPR=$1000 or HPR=1.1?
I'm sorry. Ratio of course. after_balance / prev_balance (i've updated my previous post)
I'm sorry. Ratio of course. after_balance / prev_balance (i've updated my previous post)
Then there's some problem in the reports produced by MT5.
A simple report is attached.
Let's calculate HPR:
HPR1 = 10002/10000 = 1.0002
HPR2 = 10003/10002 = ~1.00009998
AHPR = (1.0002 + 1.00009998) / 2 = 1.00014999
StdDev of HPR = Sqrt((1.0002 - 1.00014999)^2 + (1.00009998 - 1.00014999)^2) = Sqrt(0.0000000025010001 + 0,0000000025010001) = ~0.000070724820
SR = (AHPR - 1) / StdDev = 0.00014999 / 0.000070724820 = ~2.12
The report shows SR = 3.