Creating a positive IO - page 15
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
Yeah. Got it.
Let's go back: "Determine TREND by MASK (parameters attached) and start making a POSITIVE MO.
The variants of entering by trend are described here:
Yes, where shall we enter? There are 2 variants - a breakout of the previous extremum along the trend and a pullback from the previous extremum along the trend.
Are there more variants? We do not work in the flat.
Try it. Suggest new options. We'll discuss. There are no ready-made solutions, and if there are, no one will share them.
I'm not going to go real. Or rather, I will, but on futures and equities and on a different platform...
There is something strange about these statements.
Thank you. I'll give it a try.
Is the number 100 justified by something, or what...?
On the subject, creating positive IR in the future is a robust strategy based on past data. Robust means just that the positive IR will persist.)
I want to clarify robustness as the preservation of positive MO.
Positive MO is based on "bought cheaper (more expensive) - sold more expensive (cheaper)" - i.e. necessarily some directional movement that we know how to identify in the past and hope for continuation in the future. Emphasised "hopefully", but that's not the only weakness. So far no one in the thread has voiced another problem.
We have identified the direction of travel, but there is an error in this definition. We deal with NE, but for some reason we always forget about the noise component when determining the direction. And this error can behave in different ways. If it is stable - it has mo and variance at least approximately constant, and the variance is small in magnitude, then there is some hope of maintaining a positive MO of TC in the future. And if the variance of our directional error reaches 100 pips or more, and it can be easily obtained at sideways on TF higher than H1, then the positive MO will not be needed due to large drawdowns.
.
Therefore the first approximation. Positive Mo is possible if we know how to identify the direction of movement, and the error of this identification will be stable and with a small spread.
The options for trend inputs have already been described here:
"... DmitriyN:
Yes, where shall we enter? There are 2 variants - on a breakout of the previous trend extremum and on a pullback from the previous trend extremum.
Are there more variants? In the flat we do not work...".
Try them out. Suggest new ones. We'll discuss them. There are no ready-made solutions and even if there are, no one will share them.
Who has statistics on price movements after a BREAK of a "previous trend extremum" and after a rollback "from the previous trend extremum"?
Maybe it's still the same 50/50, and we're going to have to get our heads in the game...
There is something strange about these statements.
Who has statistics on price movement after a BREAK of the "previous trend extremum" and after a pullback "from the previous trend extremum"?
Maybe it's still the same 50/50, and we're going to be straining our brains here...
and how many pips are not pips? It's just that the main thing here is that tp=sl. I'm going to run with a longer profit now.
Lizanka, the main trick with mo=0.8 is always a peculiarity of quotes of that DC, on the history of which the tests were conducted. The mathematical expectation should always be several times bigger than the most aggressive spread for that currency pair. For EURUSD, you should not place non-real MO systems below 60-80 five-digit pips with a fixed volume of 0.1 lot.
Who has statistics on the price movement after a BREAK of the "previous trend extremum" and after a pullback from the "previous trend extremum"?
Maybe, it will be the same 50/50 and we will strain our brains here...
A few years ago there was a topic on the zigzag on this forum. I called it an h-zigzag for myself. It is elementary to build it. Suppose we move upwards and if the pullback occurred by more than h points, the zigzag would change its direction. Downwards, the same thing happens. We trade based on signals of the zigzag. Trading can be profitable if the average amplitude of the zigzag is more than 2h.
Before I trade on a symbol, I count the h-zigzags for the instrument at hourly bar closures on the watch. We obtain, for example, the following picture:
x-axis is a zigzag step, y-axis is the result of virtual trades of the instrument during the period of time with this step. The upper picture shows trading without the spread. The lower one - with taking the spread into account. Determine a comfortable value of the entry order and trade.
It's not 50x50
Urrraaaaah!!!
Finallyooooooo!!! NOT 50/50!!! .....
All that remains here now is to accurately define the feature sets for unmistakable identification of the event that is occurring (this is a FALSE; this is a FALSE; this is an OBSERVATION; and this is a FALSE OBSERVATION)... and we can get to work on the question of how to make sure that when we find out what's going on, it's NOT too late...