Machine learning in trading: theory, models, practice and algo-trading - page 151

 

A good portion of all scientific publications from all fields are written in R. Nature, Physical Letter Reviews, etc. The articles contain graphs made in ggplot.

There is Python, which is also very popular for analysis. What makes forex so different? The same science, there is a subject, an object, methods.

 
Alexey Burnakov:

Who has tried? My colleagues and I want to train a convolutional NS. There's some mapping going on. We hope.

It seems to be a non-standard application of the method. On the other hand, we simply present a one-dimensional "picture" as input and it's possible to smap neighboring "pixels" and their various interactions.

Convergent NS is trendy now as RF, basting etc. have been recently, but their application context is limited by pictures, in fact the only CNN trick is hierarchical selection of filters when there are obvious correlations between neighboring components of the input vector, it can be done otherwise without backprop, e.g. by clustering. But the main question is about something else: what to feed to the input.

I think I'm not revealing America especially at the end of 2016 that tn. "TA patterns", i.e. price "figures" of one series, now have no connection at all with the direction of the next tick or their average direction over a certain time horizon. It's like predicting the ball at a soccer field and using it as a predictor for the scoreboard, the correlation is weak, dozens of times less to recoup transaction costs. I think that the market is decentralized, unfortunately there is neither a standard feed nor a normal one, the ones that are available are fake. In general, to trade on the Forex ... Well, you know what I mean)))

If I'm working at quant-fund and I can talk only about obvious things, it would be stupid and illegal even to suggest how to look for effective predictors, but it's pity to see how intelligent people go through "patterns" like head and shoulders and various candlestick shifts, putting ML on it, The dependence in the price is minimal and decreases exponentially, the past price history of one series of a given length is connected with the future price history of the same length by the value beyond 2 sigmas, the deeper history makes no sense at all. Look for Causes and the fastest possible sources of data on anything that can in any way affect the behavior of the masses of people all over the planet and in all that primary broth look for precious ALPH.

 
J.B:

1) I think I'm not discovering America especially at the end of 2016 that tn. "TA patterns" i.e. price "figures" of one series, now have no connection at all with the direction of the next tick or their average direction for some time horizon.

2) It doesn't matter what you use by correlation as a search for patterns, NB, SVM, RF, MLP, CNN, etc. whatever. It's like predicting which goal the ball will hit at a soccer field, using as a predictor the sequence of score changes on the scoreboard, the connection is very weak, tens of times less to recoup transaction costs.

3) Forex is decentralized, unfortunately there is neither a normal feed nor a normal market, those that are available are fake. In general, to trade on the forex market... Well, you know what I mean)))

If my aim is to return the traded signals to the market, in order to get an idea of the market, I would use the "MetaTrader 5" as an indicator of the market situation.

1) I don't quite agree yet, I think it is quite possible to predict the price by searching for similarities in the past but it's not "TA patterns" unequivocally.... research is still underway but very time consuming in terms of mach. time

2) I agree, the model itself has little effect

For example if we take the Ribbon, separately divide it in buys and sales, then build their cumulative sum and draw the difference between these sums - you will get the same price, the sliver is a ticker, on the contrary, it predicts the price, at the highly liquid markets, the price always follows the greater liquidity, if there are more bids to buy than to sell in the sliver, then the market almost always goes down, but the difference between changes in the sliver and the price change is measured in millisecondsIf there are more bids than sales, the market almost always goes down...

I once checked the full orderlog of a symbol, many people here don't even know what it is.There is also indexation of a concrete participant in the deal and if there are 100 sell contracts in the market you can check if the only one person put this order of 100k or if this is the number of people who placed one price and it made 100k in total, you can also check if this person was taken or only 20k were poured for 100k and he cancelled the rest 80k etc.д..

So what am I saying all this, returning to the cup and the speed, there was a time that I calculated a guy who for a few ms managed to put up and cancel his request 4 times, just manipulated the price does not let it drop or rise, I have the same from the terminal only my order to poke or sell goes 0.8 sec.

4) Yes you are engaged in arbitrage in its many different forms what is there to hide :) from hft to long seasonal...

 
J.B.:

it wouldbefoolish and illegal to even hint at how to look for effective predictors

A non-disclosure signature in a quantum fund?
Or do you mean that this is a useless activity, and should not be popularized?

 
J.B:

Convergent NS are in trend now as comparatively recently were RF, boosting etc., but context of their application is limited by pictures, in fact only CNN trick is hierarchical selection of filters when there are obvious relationships between adjacent components of input vector, it can be done in another way without backprops, for example by clustering. But the main question is about something else: what to feed to the input.

I think I'm not revealing America especially at the end of 2016 that tn. "TA patterns", i.e. price "figures" of one series, now have no connection at all with the direction of the next tick or their average direction over a certain time horizon. It's like predicting the ball at a soccer field and using it as a predictor for the scoreboard, the correlation is weak, dozens of times less to recoup transaction costs. I think that the market is decentralized, unfortunately there is neither a standard feed nor a normal one, the ones that are available are fake. In general, to trade on the Forex ... Well, you know what I mean)))

If I'm working at quant-fund and I can talk only about obvious things, it would be stupid and illegal even to suggest how to look for effective predictors, but it's pity to see how intelligent people go through "patterns" like head and shoulders and various candlestick shifts, putting ML on it, The dependence in the price is minimal and decreases exponentially, the past price history of one series of a given length is connected with the future price history of the same length by the value beyond 3 sigmas, the deeper history makes no sense at all. Look for Causes and the fastest possible sources of data on anything that might in any way affect the behavior of the masses of people all over the planet and in all that primary broth, look for precious ALPH.

So according to you it doesn't make sense to trade on large TF (day, week) at all? But the big banks, successfully trade on forex exactly in the long term.
 
sibirqk:
So you think that trading on large TF (day, week) makes no sense at all? But the big banks, successfully trade on forex is in the long term.
Did the banks tell you that?
 
mytarmailS:
Did the banks tell you that?
Their public reports on trading positions.
 
J.B:


..... the past price history of one row of a given length is related to the future price history of the same length by an amount beyond 3 sigmas, the deeper history makes no sense at all.


What do you mean by that? The phrase seems to sense something interesting, but the meaning escapes me.

 
Alexey Burnakov:

What do you mean by that? There seems to be something interesting in the phrase, but the meaning escapes me.

He says what has been said here many times - the series is not stationary, the variance tends to infinity.
 
Dimitri:
He says what has been said many times here - the series is not stationary, the variance tends to infinity.

I would still like to hear his answer.

That the data is unsteady is a repeatedly stated fact.