Econometrics: one step ahead forecast - page 25

New comment
 
yosuf:
The anomaly detected should therefore be checked for this "lice" as well.
I don't think anyone in the world knows how to do that. A profitable TA system is the result of luck, inspiration and something else that is not tangible.
 
yosuf:
What prediction probability would you be satisfied with?
Look at the forecasts I make in this thread. I've laid out the formula for the prediction model. It's a simple conversion to an indicator, so what? Is it a good model? Can I go real?
 
Read the popular literature on KP. I'm going to do some reading and then I'll come back. I was more interested in this indicator's calculation of flux and prediction. I should try the forecast at least on one currency pair for now and check out the flitches. May be something will be squeezed out of this indicator)))
 
faa1947:

The market has trend, seasonality, cyclicality and noise as well as spikes (catastrophic news). Read books on market psychology, for example C-4 gave a link to a very interesting book at the beginning of this thread.


Enough of quoting books. Build a spectrum of some exchange rate and show us that cyclicality. Even if this cyclicality existed, it would be a period of 10-20 years. How would this cyclicality help you in day trading? Do you really think that trend and outliers are signs of a deterministic process?
 
gpwr:

Enough of quoting books. Construct a spectrum of some exchange rate and show us that cyclicality. Even if that cyclicality existed, it would be a period of 10-20 years. How would this cyclicality help you in day trading? Do you really think that trend and outliers are signs of a deterministic process?

Enough of quoting books.

Not quoting, but calculating specific predictions.

Construct a spectrum of some exchange rate and show us that cyclicality

There is no cyclicality in the Fourier sense.

Even if this cyclicality existed, it would be over a period of 10-20 years.

Where do you get this information, do you know how to identify cyclicality?

Do you really think that trend and outliers are signs of a deterministic process?

I have never stated such nonsense anywhere.

Don't you have anything on the subject of this thread?

 
faa1947: But here on the forum I can't get to that stage, because suddenly the model I'm using predicts correctly. But the questions still remain, as timing success in forecasting is not proof of the success of the model.

Based on 4-5 forecasts with huge errors, to draw such conclusions as "the model used predicts correctly" is unreasonable to say the least. It's just dizzying success.

P.S. By golly, you're behaving like unscrupulous colleagues from the Central Asian republics...

 
Mathemat:

Based on 4-5 forecasts with huge errors, to draw such conclusions as "the model used predicts correctly" is unreasonable to say the least. That's just dizzying success.

P.S. By golly, you are behaving like a not-so-degreed colleague from the Central Asian republics...

You don't seem to read any further than that.

Do your best to say something of substance.

 
faa1947:

In the Fourier sense, there is no cyclicality.

In what other sense can we talk about cyclicality? Is cyclical in your sense "the market goes up and then down and then back up again"?

 

faa1947: Такое впечатление, что дальше ника не читаете.

I'm reading, I'm reading. That's why I quoted two sentences, not one.

Stretch out and say something of substance.

I don't have anything substantive to say. I don't see any substance yet.

Well, I am not attracted to such models, which do not make any physical/economic sense at all.

 
Mathemat:

I read, I read. That is why I quoted two sentences instead of one.

I don't have anything substantive to say. I don't see any substance yet.

Well, I am not attracted to such models, which do not make any physical/economic sense at all.

Give me an example of "meaningfulness" and "physical/economic sense"
1...181920212223242526272829303132...139
New comment