Using neural networks in trading - page 29

New comment
 
EconModel:

Maybe I don't understand something.

We classify into patterns . We believe that such a pattern will surely emerge in the future and we will be able to use this knowledge to make predictions. Right?

On what grounds? Who has proved that there will be such a pattern at all, or slightly or strongly changed?

IMHO if we teach the net to recognise the handwritten letter "a", then there is absolute certainty that this letter will be in the future, because it exists in the language and if in the future most people start writing with their feet, there will still be an "a", just the lettering will change and perhaps the net will have to be further trained. It speaks of stationarity.

Quotations are a non-stationary process in principle, i.e. there are some kind of deviations all the time, different at different times, which are comparable to (exceed) the stationary part. This is the problem - the non-stationarity of the original: Russian letters today and Chinese letters tomorrow. One has to look for the objective reality that the letters reflect. And this is not what neural networkers do.

1. If the past is not repeated, any forecasting is doomed to failure.

2. non-stationarity and non-repeatability of the past are in no way related

3. What is "objective reality" in this sense?

 
Demi:

1. if the past is not repeatable, then any prediction is doomed to failure

2. non-stationarity and non-repeatability of the past are not related in any way

3. what is "objective reality" in this sense?

1. I disagree. The fact that you went to Kievskaya station does not imply that there will be Park Kultury next. It does, however, follow from the design of the metro. That's why you have to work on the metro system, not the stations.
2. Stationarity will not be repeated in the future, but non-stationarity will be.

3. A chair is still a chair, but the spelling of the word ..... The classification into tables and chairs has perspective, but the classification into words, by which these concepts are written, is unpromising.

 
EconModel:

1. I disagree. The fact that you went to Kievskaya station does not imply that there will be Park Kultury next. It does, however, follow from the underground structure.
2. Stationarity will not repeat in the future, but non-stationarity will.

3. A chair is still a chair, but the spelling of the word ...... The classification on tables and chairs has a perspective, but the classification on words, by which these concepts are written, has no perspective.


)))))))))) what metro? What Africa?

all existing forecasting methods are based on analysis of past process. This is what you understand as "the past repeats itself". What does non-stationarity have to do with it?

Systems for which the past never repeats, or repeats very rarely, are dealt with by chaos theory - that's not for you.

Shame on econometrics - speak more clearly, without the underground and the Chinese alphabet

 

Data preparation, noise whitening, data completeness, correct normalisation, correct teacher => profit

In fact, judging by the heat of argument on page 29, it looks like they are discussing the prospects of using neutrinos ))

 
We need to talk about something other than Navalny
 
FAGOTT:
Well, we need to talk about something other than Navalny

It's just that we need fullness of information, and people are discussing some ephemeral issues.

p.s. I've had enough of him.

 
FAGOTT:


)))))))))) what underground? What Africa?

all existing forecasting methods are based on an analysis of the past process. This is what you understand as "the past repeats itself". What does non-stationarity have to do with it?

Systems for which the past never repeats, or repeats very rarely, are dealt with by chaos theory - that's not for you.

Shame on econometrics - speak more clearly, without the underground and the Chinese alphabet

I can also be blunt: modern econometrics deals only with non-stationary random processes. Random walks have hardly been taught to me any more. This is without metros and hieroglyphics, but without them, how can I explain to proponents of neural networks that they are looking for some particulars from which no prediction can be made.
 
EconModel:
I can also be blunt: modern econometrics only deals with non-stationary random processes. Random rambling has hardly been taught to me anymore. That's without metros and hieroglyphics, but without them, how can I explain to proponents of neural networks that they are looking for some particulars from which no prediction can be made.

Errrrrr.... I thought it was only stationary processes that modern econometrics deals with! And non-stationary processes are reduced to a stationary form as a result of various manipulations.

And one should write so - I think that they are looking for some particularities from which no prediction can be made . Like, this is your IMHO.

 
I will not even hide the harsh truth from you and tell you like an artist to an artist - modern econometrics has no methods for predicting non-stationary series. Only and exclusively stationary ones and those non-stationary ones that can be reduced to a stationary form
 

We need to define the object we are working with. Where do neural networkers have this definition? What do they work with? With layers, perseptrons?

The premise: we observe the realisation of an unsteady process, usually of interest for the last 30-50 observations at most.

Then we decide what we trade. Most people trade a trend. We watch and see the trend and believe that the trend will be in the future and the past has nothing to do with it. We just believe and the past is just for model building.

This is the initial premise.

And then there are the nuances.

1...222324252627282930313233343536...40
New comment