Machine learning in trading: theory, models, practice and algo-trading - page 575
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
It can be felt. As Feynman said, a thought experiment costs nothing. You can start up a thermonuclear reactor like two fingers - what energy does it need, only 4-5 times more than in a TV kinescope.
I told you, I need NeuralNet for VisSim. I will see - what is possible and what is not.
PS I'm a theorist - I can generate ideas in high-frequency mode, but in practice - there's only 1 deal on my account, and it's a curve... Although I've started 20 pairs at once today. Let's see :)))
PPS Aha. I'll read it at my leisure. Thanks.
Your TS is strange. With 20 pairs and one trade. It's already a month you accumulate statistics, but when will we see the result?
Your TS is strange. With 20 pairs, one trade. It's been a month since you accumulate statistics, and when will we see the result?
The previous 2 weeks there were only 4 pairs and only 1 trade - yes. Overdid it with defining quantiles of confidence probability. Scared for the deposit, to put it simply. But, today I connected 20 pairs, but on demo account - we'll see.
PS Tick data I continue to take from the real account, that is, I have 2 terminals (real and demo) open on my computer. From the point of view of purity of the experiment - everything is OK.
The idea for this implementation was to find non-linear dependencies for pair trading
I finished it, drew the spread chart, looked at it with my eyes - it seemed interesting
Now I need to transfer it to the bot. I have already added an implementation of a linear spread indicator to the code base. If it will not work, I will do it too :)) But this one is much more tricky :)
The previous 2 weeks there were only 4 pairs and only 1 trade - yes. I overdid it with defining quantiles of confidence probability. Scared for the deposit, to put it simply. But, today I connected 20 pairs, but on demo account - we'll see.
Actually this is a really strange TS. Collecting ticks in millions and one deal per week.
I work only with minutes (ticks only on the last candlestick), and about 10 deals a day, on one instrument.
I don't understand it.
Actually, a really strange TS. You collect ticks in millions and one trade per week.
I only work with minutes (ticks only on the last candle), and about 10 trades a day, on one instrument.
I don't get it.
I don't know. Reassured myself... I attribute it to the trembling in my knees - fear of being embarrassed in front of the public. I wish I had gotten the results first, and then bragged about the theory... But I realize it only now.
I don't know... I'm overreliable... I put it down to trembling in the knees - fear of embarrassing myself in front of the public. I wish I'd gotten the results first, and then bragged about the theory... But, it's only now I understand.
So we're all like that here, well, maybe most of us.) First, a long talk, but comes to fruition only a few.
By the way, read the history of Student - emerged from the need to estimate distributions in small samples . I have a sample of 1 minute and a maximum of 24 hours to estimate. Imho, that's plenty enough to go around. Analysis of ticks only at the last minute.
So we're all like that here, well, maybe most of us.) First we talk a long time, but only a few of us get to the point.
By the way, read the history of Student - came out of the need to estimate distributions in small samples . I have a sample of 1 minute and a maximum of 24 hours to estimate. Imho, that's more than enough.
Yes, I have already tweaked some things.
But it doesn't change the essence of this particular branch - NS are very interesting and, remembering the thesis that one should not re-invent the wheel, but use what people have already done - I continue to look for NeuralNet Add-On for VisSim...
Actually, a really strange TS. You collect ticks in millions and one trade per week.
I only work with minutes (ticks only on the last candle), and about 10 trades a day, on one instrument.
I don't understand it.
I even have 10 trades per instrument on a 5 minute one. Ticks and minutes are not used. I'm looking to get rid of the extra ones. I have no problems with them.
I have even on the 5-minute gives 10 trades per instrument. The ticks and the minute ones are not used. I'm looking at how to get rid of the extra ones. All are good, it seems.
I've started to use fictitious profits.