From theory to practice - page 1541

New comment
 
Алексей Тарабанов:

Does it bother you?

Yes, why are all the nerds like this, not even familiar with econometrics?

trying to look for patterns in the leftovers... of their former glory.
 

It's a three-modal kind of thing.

 
Evgeniy Chumakov:

It's a three-modal kind of thing.

Head-shoulder pattern).

 
khorosh:

Head-shoulder pattern).

Not sure, maybe a diamond. The bottom is fixed.)
On the gold at the beginning of the week there was a gorgeous diamond. It worked out very well).
 
Alexander_K:


1. I've told you before. But, you were doing something wrong. There should be no drift, no MA on the chart. It should look like the bottom chart:


Construct a correlation between the increments of the upper graph and the increments of the lower graph. It floats and will only be equal to 1 if the window is taken with a fixed reference point.




for example the sum of the increments with a window of 20 minutes.


I mean if changes of price and bottom charts are not equal, then the forecast will be of the same quality.

 
Alexander_K:

1. I've told you before. But, you were doing something wrong. There should be no drift, no MA on the chart. It should look like the bottom chart:

Try building the same one on the minutes for GBPAUD for August 2019. If you get it right, we'll continue the conversation.


I did as you did, the trades for August match, and also showed test results on many pairs from 2018

 
Alexander_K:

But, Warlock has been involved in its development, his runic manuscripts, and permission has been obtained from him to distribute and use it. Obviously because he knows that converting the original BP to a stationary form is an incredibly difficult task. But whoever solves it, let him use it and make money.


I can only draw such a graph, what to do with it?



The sum at the top, the increments at the bottom.

 
Evgeniy Chumakov:


I can only draw a graph like this, what should I do with it?



The sum at the top, the increments at the bottom.

the sum of what? the increments with some kind of lag?

 
Alexander_K:

So, if real-time is able to convert the current probability density function to a normal form, the process becomes like this:

There's nothing to be gained on the upper chart with the MA. At best it is 0.

On the lower one, with constant dispersion and mathematical expectation=0, the earning algorithm is as follows: crossing of the upper line - SELL, crossing of the lower line - BUY. Exit from the trade - when it returns to 0.

This is absolutely true, and I would never have told about this algorithm, if it was mine 100%.

But, Warlock was involved in its development, his runic manuscripts, and got permission from him to distribute and use it. Obviously because he knows that converting the original BP to a stationary form is an incredibly difficult task. But whoever solves it, let him use it and make money.

Is the bottom one just the sum of the increments? or are the increments with some kind of lag co-optimised (in other words)? what is the lag then, roughly?

I'm just trying to figure out what to do with the riddled increments (that would be good)
 
Alexander_K:

It's not that.

That's it, it's your whissim that's glitchy.)


Alexander_K:

It's not. I had eight trades. Seven in the black, one in the red.

Zhenya seems to be the closest to the truth... If he also shows the constant variance.

Look at the lower chart from the beginning of the year 7+ 1-

1...153415351536153715381539154015411542154315441545154615471548...1981
New comment