From theory to practice. Part 2 - page 116

New comment
 
Roman:

Ah, well, that's different then ))

here's the MNC I just made.

why compare?

can you see where to trade? :


if you make an MNC quote - it's not like that at all ;)))))))

you buy an eva and it goes down ! ;)

the formula for the 9th grade, ahahahaha

 
Renat Akhtyamov:

Here, I just built an MNC
Why compare?
Now you can see where to trade, right?

If you take this approach, you don't need an indicator, just mark it on the initial series.
I thought you were trying to build a model that would be as adequate to the initial series as possible.
That's why I suggested that you need a comparison with the original series.
Never mind))

 
Roman:

Well, if that's the approach, you don't need an indicator, you can just add it to the initial series .
I thought you were trying to build a model that would be as adequate as possible to the initial series.
That's why I suggested that you need a comparison with the original series.
Never mind ))

masons, this is deception after deception.

expose and exploit

otherwise you will blindly believe there is a deception

;)

 
Renat Akhtyamov:

if you do the MNC quotation, it's not like that at all ;)))))))
you buy an eva and it's in the bottom, you radical ! ;)

ahahahaha

That's what it takes to sell something at the same time

ahahahaha

 
Aleksey Nikolayev:

This is not entirely true. There are symmetric dependencies that do not change when you rearrange the arguments. Besides, there can be all sorts of dependencies like "a unit occurs exactly once in a sample" - this is not typical for independent samples.

I'm talking purely about mixing the sample. The dependent and the mixed sample have the same mean.
OK, never mind, please tell me another question. Suppose we have a bimodal sample, like a sine wave or a salary at a factory) In this case, the arithmetic mean does not characterize anything at all; it does not correspond to the "conventional" mean. Is there such a criterion for applying a mean, like "the sample must be normal"? And what is it called?
 
secret:
I'm talking purely about sample mixing. The dependent and mixed samples have the same mean.
OK, whatever, please tell me another question. Suppose we have a bimodal sample, like a sine wave or a wage at a factory) In this case, the arithmetic mean does not characterise anything at all, it does not correspond to the mean "by definition". Is there such a criterion for applying a mean, like "the sample must be normal"? What is it called?

Could this note be of any help?

 
secret:
Yeah I'm purely talking about sample mixing. Dependent and shuffled have the same mean.

Roughly speaking, mixing weakens the dependence, but does not remove it completely. In fact, probabilistic dependence is the most important part of the theorem in terms of practical applications. When I was watching a youtube theorist course at MIT for engineers - it was all about it.

secret:
Suppose we have a bimodal sample, like a sine wave or wages in a factory) In this case, the arithmetic mean does not characterize anything at all, it does not correspond to the mean "by definition". Is there such a criterion for applying a mean, like "the sample must be normal"? And what is it called?

The arithmetic mean (when defined) is used regardless of non-normality, because a) the arithmetic mean of almost any non-normal converges to normal and b) it is often used as a "location" parameter for parametric families of distributions (even asymmetric ones), along with "shape" and "scale" parameters. If MO is not defined, it is changed to median.

Although maybe I just don't have experience with bipartite)

 
Roman:

Would this note be helpful?

Why do you need averages and MAs put together?

The signal is always in the centre of the trend at best.

stop averaging

Cut through all the bullshit with the AK's at least

you'll get a better deal.

;)))

 
Renat Akhtyamov:

what do you need those averages and MAs put together?
the signal is always in the centre of the trend
stop averaging
thin out all the bullshit together with AK at least

;)))

In the centre of the trend, the signal is already out of position )))
I don't want to thin out the bullshit ))
Yes, and I don't use wipers in the model.

The answer to the question was whether there is any criterion for applying the average.
The note just mentions the Shapiro-Wilk criterion.

 
Aleksey Nikolayev:

Roughly speaking, mixing weakens the addiction, but does not remove it completely.

If there was a maths thread here, it would be interesting to have a chat there.)
1...109110111112113114115116117118119120121122123...179
New comment