From theory to practice - page 1134

 
Alexander_K:

See post above.

recursion however... and by the way exponent, fractal self-similarity :-)

when the brain rests from the burden of being, it sometimes generates or perceives curious theses; which then cause either laughter or amazement

 
Alexander_K:

I don't know. Still, I have to admit that the old MQL4 forum no longer exists. There is an MQL5 forum - for programmers and signal sellers. Alas...

There was nothing special on the MQL4 forum. It was just the beginning, everything was new. It was a feeling of discovering new possibilities. False. There were already terminals with full-fledged COM-ARI for any software, from C to Excel. And in 2008 I ran away to these terminals.

What's left? - The best forum on autotrading in the Russian-speaking world. And this one is withering away.

 
Yuriy Asaulenko:

There wasn't much in the MK4 forum. It was just the beginning, everything was new. There was a sense of discovery. False. There were already terminals with full COM-ARI for any software, from C to Excel. And in 2008 I ran away to these terminals.

What's left? - The best forum for auto-trading. And this one is withering away.

Still. There was Prival, Matemat and the like. It was after reading their posts that I came here and not elsewhere. There was a desperate branch called "Predictions and Consequences..." with Strange, Tantrik, Sorcerer, where the dough was easily chopped up. Now it's only the Colonel who teaches young people how to use rails.

If we imagine that everyone was killed by the market like a child, is it worth continuing the quest for the Grail?

The answer is - yes. Because all of them are suckers and have no idea about non-linear space-time, or about anything else too :))

And we will continue our journey to the Grail. But from Monday.

Good luck to all of you!

 
Alexander_K:

All the same. There was Prival, Matemat and so on. It was after reading their posts that I came here and not elsewhere. There was a desperate branch called "Forecasts and Consequences..." with Strange, Tantrik, Sorcerer, where they easily chopped up the dough. Now it's only the Colonel who teaches young people how to use rails.

If we imagine that everyone was killed by the market like a child, is it worth continuing the quest for the Grail?

The answer is - yes. Because all of them are suckers and have no idea about non-linear space-time, or about anything else too :))

And we will continue our journey to the Grail. But from Monday.

Good luck to everyone!

That's a new one. In 2008 I perceived it like that too - I was just at the beginning. Now I'm thinking - shit... I think now that it hasn't influenced my views and trading.

What has influenced - a few answers to my questions by a famous trader on another forum. And this ideology you know)). Yes, it's from 2008. It's a win-win, especially with my hands.

 
Yuriy Asaulenko:

This is for a newcomer. Back in 2008 I felt the same way - I was just starting out. Now I'm thinking - shit... It had no effect on my trading and philosophies.

What has influenced - a few answers to my questions by a famous trader on another forum. And this ideology you know)). Yes, it's from as far back as 2008. It's a win-win, especially with my hands.

I agree - the idea itself is conceptually correct. I agree - the idea is conceptually correct and working. You just have to apply it For non-linear time and then everything will automate.

In general, if it works for me and you show me your lost deposit - I will give you the development of this idea as an active participant in the Grail Race of Sufferers.

Times and morals have changed - yes. Now it is solely and only traders, really Now, only traders in search of the Grail (and this is clearly visible in their posts), who beat like wounded lions, may claim help from my side (if the theory is confirmed by practice). The rest of them don't. They don't deserve it. That's it!

 
Alexander_K:

I agree - the idea itself is conceptually correct. And it works. You just have to apply it for non-linear time. and then everything automates.

In general, if it works for me and you show me your flush deposit - I'll give you the development of this idea as an active participant in the Grail Race of Sufferers.

Times and attitudes have changed - yes. Now it is solely and only traders, for real . Now, only traders in search of the Grail (and this is clearly visible in their posts), who beat like wounded lions, may claim help from my side (if the theory is confirmed by practice). The rest of them don't. They don't deserve it. That's it!

Thanks, I don't need it. Just wrote in the MO thread that the system set up over a year ago is still functioning fine. There's no rush for a new one. Looking for one. The problem is that it has to be qualitatively better than the old one. There is a problem with it). Every time you put it together, it always turns out to be an AKM.) Maybe that's the limit. - I don't know. All the same, we are dealing with randomness.

 
Alexander_K:

Memories of the MQL4 forum.


 
Yuriy Asaulenko:

Could this be the limit? - I don't know. We are dealing with randomness, after all.

As far as the concept goes, yes. You can't find a better one. However, there is always room for perfection within it. From my point of view, these are the nuances that increase the probability of a positive outcome:

1. working with non-linear time. How this is done - I may or may not tell you.

2. choosing a different measure of the central tendency of the process than the SMA.

3. using SL and TP.

4. Neural network as a classifier of entry points (??? don't know, haven't used it).

etc.

In other words, if you squeeze out all the details from the idea, you get a sort of worker's grail.

 
Alexander_K:

When it comes to the concept, yes. You can't find a better one. However, there is always room for perfection within it. From my perspective, these are the nuances that increase the likelihood of a positive outcome:

1. working with non-linear time. How this is done - I may or may not tell you.

2. choosing a different measure of the central tendency of the process than the SMA.

3. using SL and TP.

4. Neural network as a classifier of entry points (??? don't know, haven't used).

etc.

So, if we squeeze out all the details from the idea, we will obtain a peasant worker's grail.

1. Don't.
2. SMAs, etc., are not used.
3. SL and TP should only be dynamic, depending on the situation.
4. NS rules in this application, but the concept of application should be worked out. My NS is no more than the cherry on the cake).
 

On the subject of the dynamic window, I did a kind of test.

I took the method with the sum of increments and intervals as the basis (that was in the beginning of the topic), because of its simplicity of implementation.

I have written a self-optimizing Expert Advisor based on the following principle:

1. There is a sliding window from A to B and a step of window change of n minutes

2. The Expert Advisor analyzes (virtually trades) for the past 7200 minutes and selects the best moving window on profit.

3. trade on the best moving window for the last 7200 minutes.


The result - the result has not improved. I have taken periods from 60 minutes to 120 minutes in 5 minute steps. If the test is longer, there is no desire to waste time.