Whether there is a process whose analysis of one part does not allow predicting the next part. - page 10

 
faa1947: There will always be sections with some non-stationarity characteristics that our model does not take into account, TC will merge and we will never get the balance line as a straight line.

You are too narrow in your understanding of stationarity, just like a mathematician.

The stationarity of the law of seriesformation is also a kind of stationarity, it is just called differently. And it can definitely be used.

But the series itself will not necessarily be stationary in the sense of "stationarity of the process". And it is not necessary to detrend it at all.

 
Avals:
You don't necessarily always have to make a prediction.

If you take down all the mathematical topics in this forum whose purpose is to analyse history without reference to the purpose of using such analysis, there would be some ridiculous amount left. But here is trading, and that is always a prediction. When we enter a pose, we predict the quote's behaviour, stay and exit - always on the basis of the forecast.

If all the forum participants would correlate their posts with the objective - the forecast, the forum would be much more useful. Everything would fall away, e.g. related to DSP, which is only interested in the signal, and that's in the past. And a lot of other things. That's why I so carefully, hypertrophied the forecast on the forum.

 
faa1947:

that's why I emphasise so carefully, hypertrophied on the forum the prognosis.

detrending!!!

The word is so funny ;)))) no well really funny word "detrending" ;)))))))))

 
Mathemat:

You understand stationarity too narrowly, just like a mathematician.

It's funny, but your diploma says "mathematical engineer".

Stationarity ofthe law of seriesformation is also a kind of stationarity, it is just called differently. And it can definitely be used.

But the series itself will not necessarily be stationary in the sense of "stationarity of a process ".

Undoubtedly. I called it "stationarity of the model" above. It's just that we probably don't know the law, and we always have the model. That's why I switched above to non-stationarity of the model rather than the series. And if the non-stationarity BP didn't shake the model, that's great, but I can't make such a model, although I've progressed a lot in that direction after saying goodbye to TA.

And detrending it is not necessary at all.

I fundamentally disagree. Always, but not to throw it away, but on a shelf for future use. Let's not forget that any deterministic component "clogs up" the stochastic component and all statistics for series with a trend are questionable.

 
avtomat:

detrending!!!

The word is so funny ;)))) no well really funny word "detrending" ;)))))))))

It sure is. Synonymous with "diarrhea." Or an antonym.
 

faa, here's a function for you:

what do you say about the differentiability and predictability of this function? And tell me about its determinability!

 
avtomat:

Nothing. I won't even think about it. There's no such thing as a quote.

I'm making some kind of model. There are unsolved problems in it. Little by little, I find out what other people have solved, and there are extremely many solved, so I borrow it and find out as I go along that it has no solution at all at the moment. Considering that there are a huge number of other models, often better and more promising than mine, it's a lot to do. And simply practising mathematics has not interested me for a long time.

 
faa1947:

Nothing. I'm not even thinking about it. There's no such thing as a quote.

I'm making some kind of model

.

There are unsolved problems in it. Little by little, I find out what other people have solved, and there are extremely many solved, so I borrow it and find out as I go along that it has no solution at all at the moment. Considering that there are a huge number of other models, often better and more promising than mine, it's a lot to do. I have not been interested in mathematics for a long time.

But if you tried to find answers to these questions using this simple function as an example, you might get rid of many of your misconceptions
 
avtomat:

Nothing. I won't even consider it. There's no such thing as a quote.

I'm making some kind of model. There are unresolved problems in it. Little by little I find out what other people have solved, and there are an extremely large number of solved problems, and I borrow them and find out as I go along that there is no solution at all at the moment. Considering that there are a huge number of other models, often better and more promising than mine, it's a lot to do. I have not been interested in mathematics for a long time.

But if you tried to find answers to these questions using this simple function as an example, you might get rid of many of your misconceptions

I am suspicious of the statements of loners, I prefer a crowd, and a large one at that.

In honour of the holiday, give me the answers to your questions. I'd love to read, but it's a shame to look it up myself.

 
Happy Victory Day!!!