Where is the line between fitting and actual patterns? - page 23
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
Where is the line between fitting and real patterns?
Looking at the market we see that possibly existing patterns cannot be parametrically constant. Every system has a level of fit and a level of regularity of one or more events.
And the preponderance towards the second level is responsible for the rationality of the trading idea itself.
Thinking abstractly. Will be interested in the thoughts of others.
From another forum replied:
There are non-parametric systems and there are self-adaptants.
Don't lump them together - they are fundamentally different things. That is, you can then pile them up if you want to - diversification of systems - meta-structures of sorts... But in development - completely different things, strongly opposed to each other.
from another forum:
There are non-parametric systems and there are self-adaptants.
Don't lump them together - they are fundamentally different things. That is, you can then pile them up if you want - diversification of systems - meta-structures of sorts... But in development - completely different things, strongly opposed to each other.
There are functional surd systems and their quasi-antiparametric satellites.....
(these are friends from another galaxy who texted me)
....................
In a pile it is of course possible.
But who will clean it up later?
We would like it to be simpler, but more obvious....
We would like it simpler, but more obvious....
I've already asked for a clearer one - they sent me to search ;). Alternatively, you could look for a ready-made answer on another forum or for non-fancy interlocutors ))
I've already asked for a clearer answer - you've been sent to search ;)), alternatively, you could look for a ready-made answer on another forum or for non-fussy interlocutors ))
1) I already wrote about search in the post on the previous page.
2) Search in other forums is absolutely useless. THEY are all around.
We are being watched. Shh .... Don't you believe it? .... Here:
....... I wasn't the only one watching.
(doomfully) We can't get away from here....
There is no need to look for the line in the thread thread thread. There is no need to confuse soft with warm.
If we take the Expert Advisor as a black box having a set of parameters, the tester is just trying to fit the story. But not patterns. It is a good idea to first use the tester as a tool for developing a working system.
Let me explain the difference. 2 steps.
1. using the tester to search for patterns. This stage is not a search of necessary parameters with the help of the tester but revelation if this system works at all and where it works and what should be discarded (may be everything). Minimum parameters, rough models. I.e. first there is a search for patterns. This also applies to NS, here we decide what kind of NS and what to feed to its inputs and what not to feed.
When experimenting with tester strategies we should look at the stability of parameters to changes, it directly tells us about the found pattern.
2. using the tester optimiser to tune a running system. I.e. we already have a profitable system at a sufficient time interval, optimisation=tuning. With OOS and other things, weeding out obvious over-optimization.
There is no need to look for the line in the thread topic. There is no need to confuse the soft with the warm.
Let me explain the difference. 2 stages.
1) Thank you. But this is not the 'Traditional Approach', these are the methods of an individualist. And unlikely to be appreciated.
2)
Practically how do you implement it?
That's how I didn't understand a word of your reasoning).
By the way, where is Figar0 ? Here or there?
Sorry, guys, but I spent 72 minutes drawing a flowchart explaining everything. + Asked a specific question,
posted it at 22.01.2011 19:38 and now it's been 5 hours 55 minutes and no answers. ((
..............
P.S. I even wanted to parser the thread and build a chart of activity by dates of posts before 22.01.2011 19:38 and after.
1) Thank you. But this is not the 'Traditional Approach', these are the methods of an individualist. And unlikely to be appreciated.
So the traditional approach is to take the IACD and look for a line in the fit and real patterns? I only wrote that we should not look for patterns where they do not exist.
lasso:
Practically how do you implement this?