Article: Price forecasting with neural networks - page 2
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
If you have a prediction accuracy of 65-70%, is that enough to make money in forex? Did you get such percentages with linear regression analysis? Or by technical analysis in general (not on separate intervals, but on representative data)?
And for Forex trading I think the term "forecast" itself is very dangerous. I prefer the terms "probability of winning" and "expectation of winning".
It's a bit of a pun, but that's how it is. I won't name any more figures in %. But I like linear regression very much and use it as one of my main tools.
If I have forecast accuracy of 65-70%, is that enough to make money on forex? Did you get such percentage with linear regression analysis? Or by technical analysis in general (not on separate intervals, but on representative data)?
And for Forex trading I think that the term "forecast" itself is very dangerous. I prefer the terms "probability of winning" and "expectation of winning".
It's a bit of a pun, but that's how it is. I won't name any more figures in %. But I like linear regression very much and use it as one of my main tools.
There are more interesting metrics, e.g. how much better (or worse) your forecast is than the naive one (everything will stay as it is). With such a metric it's non-trivial to be better than a naive forecast on minute data for example. If you're interested, the formula for calculation is: U = sqrt[sum{(P_i-A_i)^2}]/sqrt{sum((A_i)^2)}, value in the interval [0,1] (Tail coefficient).
On behalf of practitioners, I would like to point out:
Forecasting the direction of a currency at 70-75% is from the realm of fiction.
I have been making such predictions for a long time, working through a bookmaker who takes bets on currency appreciation/depreciation over a fixed period of time (intraday). A rate match at the beginning and end of the period was considered a loss. At first, the bookmaker's commissions were so small that strategies with only 52% of correct predictions yielded profits. At first I used a simple system based on tehanalysis, which gave me about 54-55% winnings.
Then the bookmaker's commissions increased and I had to improve the trading system. I took all the indicators I was using and put them in . The winning percentage increased to 59-60%. So there are tasks in which neural networks rule, despite the opinions of sceptics!
If you could clarify your position: whether 75% is fantastic, or whether the networks rule...
Or do they rule up to 60%? (which is not bad either).
If the forecast accuracy is about 65-70%, is it enough for earning in Forex? Did you get such percentage with linear regression analysis? Or by technical analysis in general (not on separate intervals, but on representative data)?
In my opinion, this is a perfectly acceptable accuracy and can be used to enter real trading.
The detail is that the pluses and minuses alternate irregularly. You can inadvertently make a long series of losses (within the specified probability of success), which can destroy the deposit. To avoid this, one should correctly select the values of orders: first, the value should be a certain percentage of available funds (for example, 1-30%), and second, it should not exceed a certain value (also in %), at which even a long series of losses (for example, obtained during testing of an Expert Advisor on a long historical interval, for example, since year 99) could not kill the deposit (in the worst case or lead to a drawdown of more than half of the deposit in a normal case).
On behalf of skeptics I would like to point out:
The market is not a dynamic system.
I disagree, because a dynamical system is a system whose state changes over time according to fixed mathematical rules; the latter are usually given by equations which relate the future state of the system to the current state. Such a system is deterministic if these rules do not explicitly include an element of chance.
The weakness of this formulation is "fixed mathematical rules", but no one has yet proved otherwise, and the entire history of forecasting relies on them.
Regards, Kirillov.
- the use of grids to predict exchange rates and even the direction of exchange rates proves less effective than the use of simple classical methods of technical analysis. Predictions of relatively simple grids do not exceed 70-75%.
Forecasting the direction of the currency rate at 70-75% is from the realm of fantasy.
I have been making such predictions for a long time, working through a bookmaker who accepts bets on currency appreciation/depreciation over a fixed period of time (intraday). A rate match at the beginning and end of the period was considered a loss. At first, the bookmaker's commissions were so small that strategies with only 52% of correct predictions yielded profits. At first I used a simple system based on tehanalysis, which gave me about 54-55% winnings.
Then the bookmaker's commissions increased and I had to improve the trading system. I took all the indicators I was using and put them in .
The winning percentage increased to 59-60%. So there are tasks in whichneural networks rule, despite the opinions of sceptics!
With "winning percentage" of 60% at ~5pp spread you can live not even in Sochi, but somewhere in Hawaii... :)
The only problem in which networks rule is remembering the pattern and searching for it in the future.
And by the way, just register a bunch of indicators in an NS is not enough.
The NS require an exceptional skill from the user, to get something meaningful from the NS
It is possible only by correctly posing the question, and correctly interpreting the result.
With all the success you've had, you must be golden by now, eh? ;)
With a "winning percentage" of 60% at a spread of ~5pp you can live not even in Sochi, but somewhere in Hawaii... :)
Izvinite, chto pishu translitom. U nas tut na Hawaiii netu russkih klaviatur :)
In the beginning the spread was zero and bookmaker took a commission, which was always increasing, but nevertheless it didn't prevent me from earning money. And since March 2006 they introduced spread instead of commission, and more than forex, so I'm here now :)
And here you can see a monitoring of my accounts:
http://www.finbetting.com/modules/perfo/report.php?uid=1
The only task in which networks rule is remembering the pattern and finding it in the future.
And by the way, just putting a bunch of indicators into NS is not enough.
NS require an exceptional skill from the user, to get something meaningful from the NS
can only be properly put the question, and correctly interpreted the result.
With all this success you must be golden by now? ;)
With a "winning percentage" of 60% at ~5pp spread you can live not even in Sochi but somewhere in Hawaii... :)
Izvinite, chto pishu translitom. U nas tut na Hawaiii netu russkih klaviatur :)
In the beginning, the spread was zero and the bookmaker took a commission, which increased continuously, but nevertheless it did not interfere with my earning. But since March 2006 they have introduced spread instead of commission, and it was even bigger than Forex spread. That's why I'm here now :)
And here you can see a monitoring of my accounts:
http://www.finbetting.com/modules/perfo/report.php?uid=1
Could you tell us a bit more about your way of being a millionaire? Maybe even open a separate thread on the forum. I've seen your posts scattered around the forums. But if everything was described in one place, it would be interesting for all! Of course no one is asking for the details of the system to be divulged. Just an essay about "How I got to this point in my life" - maybe even an article here;o). If it's forbidden here, you can do it on another forum. Or give me the link again, if you've already described it all somewhere else. Thank you!