Econometrics: let's discuss the CU balance sheet. - page 24
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
i.e. trust 100%?
yes...if it's os and it works on the whole story...
faa1947, tell us a bit about yourself, where did you fall into econometrics, what do you have to do with it anyway?
Where did you fall into econometrics
Fell out of TA.
What do you have to do with it at all?
An amateur with a basic mathematical background, which I have spent my life applying to economics. In need of a hangout on econometrics for that reason.
Demi, that picture of you on pg. 13 is not stationary.
I know. I was too lazy to look for another one. Same EA, same period, smaller lot, reduce the angle of inclination and everything will be fine.
The point was that the equity line "cannot be stationary in principle".
Yes...if it's os and it works for the whole story...
And this thread questions that and proves it can't be done.
And this thread questions that and proves that it can't be done.
And this topic calls it into question and proves that you can't do that.
take m15....do a year 2002... then watch oos before 2002 and after... if the residuals are the same on both training and oos) feel free to bet... it will pull up to +...
if the residue has a normal distribution...then you have made a grail...
Take m15... spend a year or so 2002 doing TS... then look at oos before 2002 and after... if the balance is stable (same on trend and oos) don't hesitate to bet... it will work out at +...
If the balance has a normal distribution...then you have made a grail...
i hesitate to ask, but still - if the residual does not have a normal distribution, but the cp has a good yield, then what?
If the yield is positive on both "trend" (pardon the Lord) and os, but the balance is "out" (pardon the Lord) - what to do?
So far this topic has not (absolutely) proved anything. The only reason I read it so far is my interest in the original idea (hypothesis only), namely, that the type of residuals distribution after equity detrending (not necessarily a straight line), can be used (perhaps after a detailed experimental investigation.) as one of the criteria when identifying potentially OOS-resistant trading systems.
Agreed, hypothesis. Too small a sample.
It comes to mind. I addressed a similar subject once, but in a different angle.
Everything can be solved by taking the balances from the championship: both the number of deals is large and the TS are different. But then I wasn't able to get the balances into a file. If someone can do it or teach me how to do it, I will make similar calculations and see what happens.
If it's a shitty one...then all of BP's flaws will penetrate into Equi...so why analyse it then)))
I know. I was too lazy to look for another one. Same EA, same period, smaller lot, lower slope angle and everything will be fine.
The point was that the equity line "cannot be stationary in principle".
))))) That's original! What were you proving then? Pieces should be compared each to each, not two adjacent ones.
It can if it dangles around zero. If the regression line is horizontal, then the balance is stationary. But just why is that here?
fa1947, in order to apply some mathematics to reality, one must realise and feel its relation to that reality. Manipulating only formulas makes it very easy to get lost and very funny... Like Yusuf the other day with linear regression.
Even if you analyse the balance line, you will never be able to determine what the future holds. You can write an Expert Advisor that, as a result of optimization, will show excellent results, while in reality it will only fail. Using your method of analysis, you can come to the conclusion that the Expert Advisor is amazingly good.
The analysis of the balance line is only necessary to decide whether to invest in this system (or trader), to understand and understand how you will feel about it, but not to assess the future. This analysis is done with the assumption that the same trading style will continue in the future. The only thing to do is to calculate the deviations in order to estimate the maximal possible surprises (drawdowns, etc.).
Expert Advisor should not just have a balance line after optimization, but a balance line from trading on the forward periods. What you are doing now is an indicator of how well the Expert Advisor has optimized.
Maybe you are really good at econometrics etc. but the fact is, you are applying it in the wrong place.