Machine learning in trading: theory, models, practice and algo-trading - page 995
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
In order not to be hopelessly sad, I once again attach Kolmogorov's work on BP forecasting.
From it, it is simply obvious that we should work with the first and second price differences. And if the expectation of returnees is always strictly = 0, then with the second differences we need to work - to do such a tricky (according to Aleshenka - exponential) sampling, so that the ACF took some tricky form (I was not familiar with it).
And voila - the Grail was taken from crying Alyoshenka and distributed to the suffering people.
No. It's better this way:
voila - the Grail is taken away from the crying people and given to the crying people.
No. This is better:
voila - the Grail was taken from the suffering Alyoshenka and distributed to the weeping people.
That's right, it's almost 50/50, there's almost nothing to trade anymore, the market is getting more efficient, the MO's exhausts are approaching the contrived indicator-based TSs of 5 years ago. That's the truth of life.
There was no grail, there was hard, persistent work and small successes at the end, which could be explained by good luck, luck with the type of market, and now sadness, I have to think about corporate espionage and insider trading in general, if I plan to continue, there is no way out soon.
The only thing that is not a shame at least, is that I spent those 7 years incredibly interesting, I had learned a lot and during the last 2 years I got back everything I had lost during the first 5 years and i am now ~12% in profit))) Although I probably stayed on my own, given the inflation of the buck.
Alesha, I understand everything, the work is hard and I do not want to give away the acquired knowledge.
I ask only one thing - write something like: "it is necessary to work with the first and second (third, ...) differences of a price range. Period."
So that people do not rush from side to side, and purposefully worked with it.
Well, it's a shame - an interesting branch is dying, no one has the result.
The only thing that at least does not hurt that I spent these seven years incredibly interesting, I learned a lot
Well it is definitely better than sitting under the stultifying effects of TV, beer, etc. (who - to what - is inclined).
I, too, have been interested for a year, though without material return.
Well, the insider, can only afford very rich people/companies. No one (who owns it) will not leak valuable information for a pittance, while risking criminal liability.
Effective, but not completely, you can predict, even almost statistically reliable, but taking into account the trade costs and taxes will be little left, with reasonable trade, that is, with adequate risk, for example, to earn as a good Moscow coder (~ 50-70k $ in code), you need to work with 0.5-0.7m $ capital itself (taking all the profits). If we talk about the hard-earned 10-100k $ profit is not enough for a comfortable life, and if you increase the risk that you can go to zero and the music will stop before it starts.
Better try to forecast for example with XGB {R(t+w,w)/sqrt(w)} by {R(t,w)/sqrt(w),...,R(t,k^N*w)/sqrt(k^N*w)} where R(t) = (price(t) - price(t-1))price(t-1), w - window, N - number of features
Then average Ema because it is very noisy, you can defeat spread, but it is not a grail of course
Thank you.
You can not do without it, then it will be strictly minus, once again, there are patterns, but which are obvious, in a competitive environment immediately compensated by colleagues, regulation, trade costs, taxes, etc.
MO now nobody will be surprised, no matter how it may seem at first, but in fact you can reach almost the limit in MO quite quickly(2-3 years), in the context of the market, further data, they are proportional to the depth of the purse
IMHO age is now in advanced NLP to analyze large volumes of textual content on the internet, at least to read and make sense of most news, the NLP we have now is very far from human. Prices are driven by news, both strong, like speeches of politicians, regulatory acts, etc. to generally social trends, which are discussed in social networks by millions of tweets, photos, etc. If you can make sense of it all and don't be stupid enough to put code on a github, you can make billions.
That's right, it's almost 50/50, there's almost nothing to trade anymore, the market is getting more efficient, the MO's exhausts are approaching the contrived indicator-based TSs of 5 years ago. That's the truth of life.
There was no grail, there was hard, persistent work and small successes at the end, which could be explained by good luck and luck with the type of market, and now it is sad, I have to think about corporate espionage and insider trading in general, if I plan to go on, I will not succeed without it soon.
And yet.
Now it is fashionable, due to small profits from trading, to trade by signals.
Why none of the old timers (Warlock, Toxic, and others, including you, of course) do not do it? I've never even seen a report from MT on a real trade.
Is there no desire to just show off your skills and get some well-deserved applause?