Machine learning in trading: theory, models, practice and algo-trading - page 918
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
yes, exactly
Do you manage to build strategies on the former predictors, how many of them do you have in units and the total number of combinations?
There are 40 of them now, but there is essentially one, just training with different conditions, then averaging the results over all of them
That's a lot, I can't even think of a hundred! And how many predictors and their number of combinations?
predictors (inputs) from 25 to 200 for different strategies :) no need to make up, just small modifications of some initial
I wonder, how do you mark their belonging to each TC? I.e. columns 200, somewhere they are filled with 25 values, and what is in the remaining 175 - a unique value of the variable (predictor)?
I shot off.
I wanted to shoot till spring, but I got it till summer.
Wait for pictures from Australia :)
Investors - consider from $ 50k
I do not understand, taught NS to earn?))
I don't get it, did they teach NS how to make money?)
Well, sort of, did almost everything I could :) not a grail, but something will sometimes work, then there is no sense to worry
Will it make 100% of annual profits without losing it? What kind of protection is there against losing? I am struggling with preserving my deposit. In my opinion, this is the key to making any expert profitable. I am serious :)
But on the whole, I support you. I have invested so much time and effort. It's time to start demanding profit from the NS.
Will it make 100% of annual profits without losing it? In general, what kind of protection is there against the drain?
Yes, that's probably the average result, taking into account periodic retraining?
If you choose persistent markets and shut them down for all sorts of bullshit news, then maybe more
stop, well.
But on the whole, I support you. So much time and effort to invest. It's time to demand a return from the NS.
The time and effort - it's not really worth it) but in terms of gaining some knowledge on neural networks, it's okay.)
I have invented reversal of emergency order (with increased volume, which leads to break-even). It seems to fight for the deposit, but not very stable. The market beats it on flat markets, where the emergency order lacks volatility on bars. Of course, it can be done in 100 iterations but the reversal is expensive. If the price passes from the emergency order open price by 10 points (the step is set) in the wrong direction, the emergency order is closed and a new one is opened in its place. The task is simple - to catch the following price direction, the price moves up and down inside the bar before the bar goes up or down... But in general it's better than just one emergency order, although the reverse sometimes gives more losses than if the EA had closed one emergency order if the drawdown increased after it was opened... However, the point is that if the repeated emergency order is unsuccessful, the deposit will suffer losses and the reverse allows the deposit to stay afloat longer. I am still testing it and reversing is implemented in case of large deposits starting from EUR 500. Reverse is relevant for large deposits - 500 EUR, but reversing eats small ones and looks perplexed, like where is the money?)
Yes, that's probably about the average, taking into account periodic retraining
if you choose persistent markets and shut them down at all the news stuff then maybe more
stop, well.
X2 for the year with minimal risk of loss of investment - it's a success!
All the rest is a plum game, in which only the date of this plum remains unknown. This is all for gamblers, and serious businessmen like stable income, and for them the one who made the % return of X2, X3 from the rate of bank deposit is a cool financier.