Using Neural Networks in Trading. - page 16
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
Smart people gave up predicting prices back in the mid-90s. So if you're predicting prices, give it up quickly, it's a pointless exercise.
What do smart people do now?
Yes
flag in your hand, I'm happy for you! :)
>>What are you approximating with, if it's not a secret?
Studying bid-ask spreads is a fairly new issue in financial mathematics, it is being studied very actively, I know financial firms in America that have made MTS on this basis, but for options (based on the spread). In forex we don't have a market rate and what is spread is unclear, but I think we can come up with an analogue based on ticks. Financial maths files attached, the rest of the science development can be found on google.
what do smart people do now?
Smart people are pumping oil. ))))
The rest of us do it by the direction of the movement or at most by the magnitude of the increment.
Smart people - pumping oil.))))
)))) unfortunately I am not that smart yet...
about the direction - it's the same thing...
Well, if you think it's the same thing, then go for it! )))
about the referral - it's the same thing...
Generally speaking, of course not. Price and direction are different by conventional standards in this case...
With neural networks, the main thing is not what to predict, but what to input. And here the science is silent, tools are invented, but there is no mathematical justification.
Well, if you think it's the same thing, then go for it! )))
no, wait a minute... i'm not predicting price levels... I'm predicting the most likely price movement at some point... Anyway, Leonid, maybe I'm missing something (that's how neural networks got started), but what roughly should what you call a directional prediction look like? what is this prediction based on?
no, wait... I'm not predicting price levels... I'm predicting the most likely price movement at some point... Anyway, Leonid, maybe I'm missing something (that's how neural networks got started), but what roughly should what you call a direction prediction look like? what is the basis for such a prediction?
Well then you're not expressing it correctly. Predicting a price is saying that the price will be 1.3567 in some time, for example. The direction is whether the price will go up or down at this point in time. All these predictions are based on the data you are using, it doesn't matter if it is a neural network or a standard TS.
yeah... I thought... Kotelnikov's theorem is more complicated... but that's for Prival to ask...
No, it's not more complicated... at first I thought that the gaps after the weekend broke the continuity... they may violate the continuity of the "real" instrument price (which also exists at the weekend), but they don't violate the continuity of my dealer's price...