Machine learning in trading: theory, models, practice and algo-trading - page 1149
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, in fact, are Koldun, Doc, Fa, Aleshenka the son, etc.? Where is Misha the teacher, in the end?
The most beautiful branch is frozen in anabiosis... What happened?
The teacher's in the car service changing carburetors, Doc drained his bitcoins, Koldun's gone to his sweetheart, FA's singing a requiem for the garbage, Aleshenka can't find a book on tics...
What's not clear?
The teacher is in the car service changing carburetors, Doc drained his bitcoins, Koldun went to live with his sweetheart, FA is singing a requiem for the garbage, Aleshenka can't find a book on tics
♪ what's not to understand? ♪
:))))
I have online training, what am I going to do with these files? I will have to synchronize the time and other hassle.
I should write everything in MT5 from the beginning, so there would be no mishaps
Man, this sucks!
man, that sucks!
Yes, I have a different approach, it's a science to shove these lefty predictors there, without coordination with the target makes no sense... look at the last libs that I added to the codebase, for example
maybe someone will do who likes it... but the last one who stayed, pan straight, was bannedMax, tell me this, because I'm already exhausted....
Is it possible to determine the beginning of the trend with a neural network? I.e. the beginning of BP section with known positive correlation of increments?
I'm tired of looking for the parameter "trend/flit"... I have an idea to connect NS as a classifier to my TS, and everything will be alright. А?
Max, tell me this, because I'm already exhausted....
Is it possible to determine the beginning of the trend with a neural network? I.e. the beginning of BP section with known positive correlation of increments?
I'm tired of looking for the parameter "trend/flit"... I have an idea to connect NS as a classifier to my TS, and everything will be alright. А?
Trend / volatilitymytarmailS did it through the acf
you don't need to search for something... Read a book about MSUA, and you'll understand if you need it or not...
http://www.gmdh.net/articles/theory/zmInduct.htmtrend/flat is whatmytarmailS was doing through the acf
You should read the book on MSUA and you'll understand if you need it or not...
http://www.gmdh.net/articles/theory/zmInduct.htmHe doesn't know a thing, in my opinion...
If you think up how to do it, I'll send you a detailed algorithm of my TS, maybe I'll be able to make a symbiosis...
He doesn't know anything, in my opinion...
If you think - how to do it, I'll send you a detailed algorithm of my TS - maybe we can make a symbiosis...
it's all the same, it's either a lagged detection or an optimization problem
in short, just use hyperparameters of the TS
it is still either a lagged definition or an optimization problem
In short, it's a dumb way to optimize the hyperparameters of the TS.
Think about it, please. Not right now - when I have time, I'm not rushing. There are many wonders in the distribution of increments - up to hyperextension and hyperasymmetry. Maybe NS can detect something that is the key? I don't know... No matter how long it takes - if you come across it, let me know. OK?
Think about it, please. Not right now, but as soon as there is time, I do not hurry. There are many wonders in the distribution of increments, up to hyperextension and hypersymmetry. Maybe NS can detect something that is the key? I don't know... No matter how long it takes - if you come across it, let me know. OKAY?
Yes, ok, I promise to think about it at my leisure ))
I use increments, but in optimization problem... it's like it optimizes itself, adjusts for history. Of course it works strangely at the moment, I haven't done anything normal myself yet.