Machine learning in trading: theory, models, practice and algo-trading - page 174
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
Swaps, in theory, should not be, because there is no supply of currency. And also interesting is the recent situation with GBPUSD. How did the futures behave, and how low did they fall? :)
So it says in the table, EXPECT a reversal, for example today on the pound.
Profitable models are obtained not even on 10000 and not on 500 predictors, but less, say, on 10. Cramming 500 predictors into NS is a bummer. Noises will drown everything out there. I'm not surprised that it didn't work well.
90% of the problem (the other 10% is to form predictor candidates and choose the MO method) is unbiased model selection. Somehow the narrow throat is often how one interprets the training results.
I'm sure that even with lots and lots of information, screwing up an experiment can be very easy.
Here you have praised Reshetov, never having understood his work, and yet he did a MEGA cool thing, Tell me why????
He solved one of the important problems in the construction of NS. It's choosing those predicates that give the most generalization. I have an unloading file has about 40 predicates, and a third of the main, the rest of the lag from these data, but the optimizer builds models using only 4-5 predicates. And they are always different. Models that used more than 5 predicates as practice has shown work not very well, a lot of values was "I don't know", like this model is seldom but accurate, and works longer, and considering that I optimize the model every day, I do not need it, Well, the number of records I do for the last 5 days (in accordance with the volume and OI) as a result, I have a table of 45 columns and 30 rows and it is enough to earn (as shown in the screenshot) and it does not matter how the network divides the good and bad, what is important that she did it steadily. Not infrequently I have to flip the TS, because after training it starts to STABILIZE to drain, but flip it and voila, we start to earn steadily, so it's like this....
but how to separate a model trained on noise from a model trained on signal.
I know how to do it, but I need a lot of deductions
Well, you're thinking in the right direction :)
Profitable models are obtained not even on 10000 and not on 500 predictors, but less, say, on 10. Cramming 500 predictors into NS is a bummer. Noises will drown everything out there. I am not surprised that it did not work well.
You should be surprised :) It worked fine at that time, I dare not talk about it now, because I'm forced to, but I "heard from someone who heard" about quants from the fund, where they give 10k vectors to CNN inputs, I don't know their recent returnees, but in 2011 they had 12% on half a yard, which is cool, although they dropped by 8% in the middle, but still...
No comment about "profitable models" on 10 features)) All who really cuts the market favorable "guru" convincingly argue that the model should be as simple as possible, that not to be overtrained, that on the mash and BB, you can build a "profitable model" and so on. Many thanks to them))
I take volume and OI from pound futures with CME, I also take real volume in real time from delta cluster. So everything is available, and the difference between futures and spot is only a few pips, so.....
Is it free or by subscription? If it is not a secret, could you tell me how in MT or Quick to get OI from futures with CME and the real volume.
I have never been used to trade on this market and I have never been used to buy from other Asian traders.
Is it free or by subscription? If it is not a secret, could you tell me how in MT or Quick to get OI from futures with CME and the real volume.
Thank you.