Econometrics: one step ahead forecast - page 45

New comment
 
faa1947:
Here's the textbook, I don't see a single item that got a Nobel Prize. Can you give me a hint?

I have a profit formula, can anyone show me a similar one, but I am not counting on a Nobel as they (formulas) will be understood later.
 
Reshetov:

It is much easier to come up with a new theory. From a practical point of view, it is likely to be of no use either, especially if cause and effect are mixed up, then to illiterate people it will look quite "plausible" on the surface, like Yusufkhoji's, for example. But the author of yet another nonsense automatically becomes the leader of a sect.

But what's the fun in it? Someone there received a Nobel Prize for econometrics and he does not care whether the theory is correct or not. And someone else, having believed in this very theoretical nonsense, now only screws up.

OK, Yusufhoja has come up with, as you put it "then to illiterate people outwardly everything will look quite 'plausible'," let literate people point out my mistakes. I am not giving up my vision of the problem, I want to see someone who is able to fight with me with his theory, albeit badly or well, would be able to predict the future as my theory, albeit not universally accepted.
 
faa1947:
Here's the textbook, I don't see a single item that got a Nobel for. Can you tell me?
However, however, in all recent years, the Nobel Prizes in economics have been exclusively in ECONOMETRICS, including Engel.
 
faa1947:

What you refer to as new directions I paid for these directions 30 or 40 years ago. In the USSR, artificial intelligence and pattern recognition were extremely well developed. I am not interested in it.

The lack of interest among the participants of this forum is explained not by the obsolete parametric methods, but by the most common crude ignorance and a desire to show off. And it is done by reinventing another bicycle.

Flip through this thread and you will see that there is no constructive criticism of what I lay out, as well as no suggestions for the development of the semiproduct I laid out in the thread.

Now, a couple of pages ago we stopped before defining the stability of the model. Maybe non-parametric econometrics can solve this question? What should the model be so that we can trust the prediction? Just don't do the forward test.

Or broader. What specific problems not solved by parametric methods can at least some non-parametric method solve.


I have been following your research for a long time and closely. You are a true econometrician, which is why you are repeating the common mistake of fitting the trading process to assumed assumptions and known patterns. Try to look at the process of trading, specifically forex trading, from a different perspective, putting aside the weight of past econometric knowledge. Finally find the function responsible for this process and develop it. I've identified it as a Gamma function, so tell me your opinion.
 
yosuf:
I have been following your research for a long time and closely. You are a true econometrician, which is why you are repeating the common mistake of fitting the trading process to assumed assumptions and known patterns. Try to look at the process of trading, specifically forex trading, from a different perspective, putting aside the weight of past econometric knowledge. Finally find the function responsible for this process and develop it. I have defined it as a Gamma function, you may say your piece.
I am a down-to-earth person. I am trying to solve specific problems at the moment - the stability of the model. I have shown in this thread and in the articles that the building blocks of model stability are: requirements for model parameters, absence of autocorrelation in the residual, stationarity of the residual. Is that all? Was hoping to get some advice from the members of the topic. Perhaps a gamma function. But where does it belong in this list?
 
yosuf: Finally, find the function that is responsible for this process and develop it. I've identified it as a Gamma function, you have your say.
Yusuf, you are out of your depth here, to put it mildly. The key point of this thread is statistics. I haven't heard a single clear word from you about statistical methods - even though you say you teach econometrics.
 

faa1947:

What should be the model to be able to trust the prediction? Just don't do a forward test.

We should, Fedya, we should.

faa1947:

What specific problems that parametric methods cannot solve can be solved by at least some non-parametric method.

The solution to the problem you mentioned above:

faa1947:
Or come up with another model within the framework of the theory. now we observe - a limited number of theories and an unlimited number of models.
There is no need to invent anything, the bicycle was invented a long time ago, because . The data itself forms the model.
 
faa1947:

What you refer to as new directions I paid for these directions 30 or 40 years ago. In the USSR, artificial intelligence and pattern recognition were extremely developed. I am not interested in it.

The lack of interest among the participants of this forum is explained not by the obsolete parametric methods, but by the most common crude ignorance and a desire to show off. And it is done by reinventing the wheel.


.

On the contrary, I am still extremely interested in fields related to artificial intelligence and pattern recognition.

.

"As a true parallel to the evolutionary development of organisms, one might consider a historical exhibition of the bicycle, which clearly shows the changes that the machine has undergone year after year, decade after decade; the same can be said of steam locomotives (diesel engines), cars, aeroplanes, typewriters and others. Here, as in the case of natural processes, the importance of constant use of a particular machine is evident, the result of which is the improvement of the latter; improvement through not literal use, but through accumulated experience and suggested changes."

(Erwin Schrödinger. Mind and Matter.)

 
Mathemat:
Yusuf, you are off the mark here, to put it mildly. The key point of this thread is statistics. I haven't heard a single clear word from you about statistical methods - even though you say that you teach econometrics.
Indeed, I am sure that statistical methods cannot describe a dynamic market, not even one step ahead. (18) describes the past history perfectly and at the same time is forward-looking, I am interested exactly in this property. Existing statistics (18) describes better than any model, and that's why I'm not testing its capabilities in static part, i.e. in describing history. The success or failure of (18) can simultaneously indicate whether the patterns which perfectly describe history are capable of predicting the future? Or are we in principle not meant to predict on the basis of historical data, as many participants are convinced? For what it's worth, and (18) in its predictions is based on historical data like any other such regression model, although I'm sure it's impossible to create a model not based on historical data for the market because of its and its extreme complexity.
 
Reshetov:



There is no need to invent anything, the wheel was invented long ago, because . the data forms the model itself.

Couldn't you be more specific?

We have to, Fedya, we have to.

Your forward test, translated into econometric language, is a breakpoint test. But it is only one of the stability tests and it is conducted in a much more diverse way than your forward tests. There is, however, a certain basis for it.

The solution to your above mentioned problem:

With an example please. On what basis may we trust the NS in terms of prognosis?

There is no need to invent anything, the wheel was invented long time ago, because the data form the model itself.

What forms the NS? It is impossible to understand at all what it has formed there. It has formed a black box which we believe sacredly.

1...383940414243444546474849505152...139
New comment