Machine learning in trading: theory, models, practice and algo-trading - page 1153
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
Is devtools installed?
Yes, without it there was another error.
Yes, without it there was another error.
Man, I don't know(( I just tried again, it all installed. I also have R-3.5.0
Damn, I do not know(( just tried again, I have everything installed. I also have R-3.5.0
I can download the file separately, but how do I install it?
I'm not quite sure why... There are "future selection" algorithms that solve the problem of separating useful predictors from noise
I'm a little bit not about that, the question rather how to implement heterogeneous predictions in the system, there are different predictors, noisy, predicting some exotic statistics, etc. but can be useful, such as "reversals" as you have)))
PS with algorithms "future selection" also not so simple, for non-linear classifications/regressions, one thing is to play with ready-made library on model data, another thing is to write a library yourself and experiment with real data.
I can download the file separately, but how do I install it?
If p-studio
I'm not about that, the question is rather how to implement heterogeneous predictions into the system, there are different predicates, noisy, predicting some exotic statistics, etc., but which can be useful, such as "reversals" as you have)))
PS with algorithms "future selection" also not so simple, for non-linear classifications/regressions, one thing is to play with ready-made library on model data, another thing is to write a library yourself and experiment with real data.
I will only say what I think and in what direction I try to dig. I think that first of all the problem of fractality should be solved, that is, roughly speaking, we should teach (the algorithm-MO TS) to see "head and shoulders" both on the daily and on the 1-minute chart, and that (the algorithm-MO TS) would understand that it is one and the same.
Then the heterogeneity in the data will disappear and stationarity will appear, and most importantly, repeatability that is absent in the raw BP, which is not useful to include in the MO
========
Or work with attributes which do not mutate in time and do not change their structure and are stationary.
Mmm-yes gentlemen... mm-hmm... 1153 pages of demagoguery! Oh, my gosh! Read for a month (mostly diagonally), the only thing I noticed constructive in https://www.mql5.com/ru/articles/1165
This whole discussion reminds me of https://www.mql5.com/ru/forum/4956 I think it would be just as useful
The only place where I noticed the constructive in https://www.mql5.com/ru/articles/1165
)))))))
Mmm-yes gentlemen... Mm-hmm... 1153 pages of demagoguery! Oh, my gosh! I've been reading for a month (mostly diagonally), the only thing I noticed constructive in https://www.mql5.com/ru/articles/1165
ZZ as a Target - facespalm
But yes, it is constructive.
I think that first of all it is necessary to solve the problem of fractality, or roughly speaking to teach (algorithm-MO-TS) to see "head and shoulders" both on the daily and on the minute chart and that (algorithm-MO-TS) would understand that it is one and the same.
Then the heterogeneity in the data will disappear and stationarity will appear, and most importantly, repeatability that is absent in the raw BP, which is not useful to include in the MO
========
Or work with attributes that do not mutate in time and do not change their structure.
Vector forecasting overcomes "fractality", i.e. we make increments for different time scales, for example for 10 minutes forward, hour, 6 hours, day, week
the "head and shoulders" is a price pattern, every algotrader has written a pettern correlator to make sure the market does not see patterns, there is no statistical reasoning that a certain "figure" can predict something, they are "visible" just like "animals" in clouds, this is pareidolia