Matstat Econometrics Matan - page 29

 
Igor Makanu:

on the physical level - yes, only mathematical models describing physical processes

on the transport layer, the matstat is unnecessary, hard algorithms are hardwired into the microcontrollers

Network card can't do much, probably more correctly as an example - ISP linking equipment or SDH

A network card, a router in any case has intelligent algorithms, albeit hardwired into the hardware. In the price range we don't know what price is next, in the network we don't know how many points will be connected, but we know the algorithm for allocating addresses. There is something in common in the tasks, but they are different in essence. Although they have the same, there are areas of situations without solutions. )

 
Igor Makanu:

on the physical level - yes, only mathematical models describing physical processes

on the transport layer, the matstat is unnecessary, hard algorithms are hardwired into the microcontrollers

The network card can't do much, probably more correctly as an example, the ISP's circuit forming equipment or SDH

I was referring to the Viterbi algorithm used in wi-fi.

And in general, modern communication theory begins with Shannon's work, which refers to theorists and matstats.

 
Aleksey Nikolayev:

I was referring to the Viterbi algorithm used in wi-fi.

Still, I have a slant in my thinking). Routes and recognition of weak signals in broadband are really different tasks))) And the second one is more complicated by an order of magnitude))

 
Valeriy Yastremskiy:

On R?

On mql.
In all seriousness, "10 lines" is just a trading algorithm. But in order to fight broker and Metatrader bugfixes, of course I had to add a couple thousand more lines to it))) And then some more to compile in 4/5-ka without code rework.
 
Aleksey Nikolayev:

Randomness in the theorem is not defined as a concept at all, but is simply used as part of the terms. Therefore, reasoning about randomness as a particular concept is usually inherent to people who are ignorant of the theorem and matstat.

Randomness = impossibility of accurate prediction, no? It seems to be quite a concrete definition, what's wrong with it?

 
secret:

Randomness = inability to predict accurately, isn't it? Seems like a pretty concrete definition, what's wrong with that?

Here's an excellent lecture on the subject


 
secret:

Randomness = inability to predict accurately, isn't it? It seems to be a very concrete definition, why is it not good?

It's OK for philosophy, but not so much for mathematics, because it doesn't give a basis for any calculations)

 
Igor Makanu:

here's a great lecture on the subject


Shiryaev has a tough language. To understand him, you need to know a little less than he does.)

It seems to me more constructive to proceed from the general concept of uncertainty (or more narrowly, uncertainty from game theory). Then one should specify when and how the general uncertainty can be reduced to its probabilistic version.

In general, I prefer to use the word "uncertainty" instead of "randomness". In the theorist, the word "randomness" is the approximate analogue of the word "marine" in the name of the animal "guinea pig")

 
Aleksey Nikolayev:

OK for philosophy, but not so much for mathematics, as it does not provide a basis for any calculations)

You have given some kind of philosophical answer)
 
secret:
You have given some kind of philosophical answer)

If you somehow define your"impossibility of exact prediction" in terms of numbers, functions and sets (like in Kolmogorov's axiomatics of probability), then you can shift from the field of philosophy to the field of mathematics).

It may well be said that accurate prediction is never possible in principle - "Trust, in our time, cannot be trusted by anyone, not even by oneself.")

Reason: