Sharpe ratio signal - page 3

 
Enrique Dangeroux #:

Not sure what is not to understand.

So, the value is relative. There is no such thing as above 1 or 10. One can consider a value of 0.2 as better then 0.1. That is all.

And like i wrote, a sharpe ratio can look better if no trades are closed with a loss, till the margin call comes off course.

To evaluate a strategy sharpe alone is useless.

I think you haven't looked for anything, because as soon as you try to know how to analyse the sharpe ration you will systematically have the explanations which tell you that a sharpe ratio lower than 1 is that it doesn't there is any security! ratio between 1 and 2 is not profitable and it takes a ratio greater than 2 to consider using the system...

And if, the way to calculate a ratio of sharpe makes a ratio lower than 1 is beautiful is well without any protection....

Only all the signals on mql5 are less than 1, I think that in these signals there are some which are good and which must surely have a sharpe ratio different from that displayed.

 

What is on the screenshot is not correct. Written by an ignorant.

As long as you do not think for yourself and consider what is written somewhere on the internet as the holy truth, you will never understand.


The yellow part is total BS.


Example sharpe ratio 0.71:



Sharpe ratio 0.07:


It is obvious the signal with the lower sharpe ratio has a better return vs risk ratio, it just does not reflect in the sharpe because the signal with higher ratio does not close losing trades.

 
Enrique Dangeroux #:

What is on the screenshot is not correct. Written by an ignorant.

As long as you do not think for yourself and consider what is written somewhere on the internet as the holy truth, you will never understand.


The yellow part is total BS.


Example sharpe ratio 0.71:



Sharpe ratio 0.07:


It is obvious the signal with the lower sharpe ratio has a better return vs risk ratio, it just does not reflect in the sharpe because the signal with higher ratio does not close losing trades.

For me in your 2 examples the ratio is not right, not because your first example does not close losing positions, but although the ratio should be higher if the calculation is done, as we know.

We are really talking about the sharpe ratio, there is no criterion in value to analyse everything, it is thanks to a set of criteria that we can judge a trading system.

We're not here either to say everything you need to know to make a choice, what to pay attention to, etc.

You even fail to recognise it and do your own research or apply. But if you take the time to document yourself, I think you will find that the sharpe ratio values ​​in the signals are wrong.

So I don't know if it's wrong? Can be a calculation that the developers of the site must explain to us. You can't make comparisons either, when you analyse a robot, you don't compare, you only analyse the trading system.

 
Enrique Dangeroux #:

What is on the screenshot is not correct. Written by an ignorant.

As long as you do not think for yourself and consider what is written somewhere on the internet as the holy truth, you will never understand.

Example sharpe ratio 0.71:

..

Sharpe ratio 0.07:

It is obvious the signal with the lower sharpe ratio has a better return vs risk ratio, it just does not reflect in the sharpe because the signal with higher ratio does not close losing trades.

Don't compare apples with oranges! As the Sharp Ratio does not consider the equity curve one cannot blame it for that but has to use a different calculation, which is definitely recommended!

But there is another aspect of Sharp Ratio, which is very interesting especially for the signals, but which does not (I think) correspond to the traditional interpretation and intention.
If, with the same average of the profits of the individual positions, their range of variation is larger, the std. dev. is larger and thus Sharp Ratio becomes smaller. Therefore, in a way, it measures the straightness of the capital curve. I have attached a small Excel sheet that shows this.

Files:
 
Please read carefully. My point was; The article states "the sharpe reflects the ratio between returns and risk". This is incorrect. No risk controll == infinite risk.
 
Carl Schreiber #:

Don't compare apples with oranges! As the Sharp Ratio does not consider the equity curve one cannot blame it for that but has to use a different calculation, which is definitely recommended!

But there is another aspect of Sharp Ratio, which is very interesting especially for the signals, but which does not (I think) correspond to the traditional interpretation and intention.
If, with the same average of the profits of the individual positions, their range of variation is larger, the std. dev. is larger and thus Sharp Ratio becomes smaller. Therefore, in a way, it measures the straightness of the capital curve. I have attached a small Excel sheet that shows this.

Very interesting your analysis, I did some tests with the excel file, and indeed we are getting closer to it I tried to add the commission and the swap but I have less difference.

Still, the sharpe ratio is an indicator of risk and if we follow the logic in the excel it is not taken into account at all. Moreover, I haven't seen a formula where you divide the average profit with the standard deviation to obtain a sharpe ratio.

 

As we are already talking about the statistics of he signals ..

The quite successful signal from above shows this:

Max deposit load:    125.93%

How is this to interpreted? Of course the max load has become 25.93% below the deposit - I guess initial deposit.

But what is the information value of this figure if the account balance has become a multiple of the initial deposit and the signal is now trading at higher volumes?

 

Is it possible the  MLQ calculation is off a decimal? Did MLQ place a bad formula? 

Reason: