From theory to practice - page 427

 
Evgeniy Chumakov:


More of an extension, the chart should be read from right to left ..... yes, not like the Russians ) You just need to write backwards into the file.

:)))) Cool. Nah, well, obviously applying the parameter from the 3rd chart will work on the minutes as well.

Voila.

 
Alexander_K2:

:)))) Cool. Nah, well, obviously applying the parameter from the 3rd chart will work on the minutes as well.

Voila.


Maybe a spike due to holes in the history.

 
 
Alexander_K2:

The price is correct, the variance is not - I don't understand why it has dropped so drastically.

Think, mathematician, think. Let's try it three times.
 
Evgeniy Chumakov:


So if the increments are dependent, is there a chance? On the subject of joking, no . Took a live reading from the graph.

One can try, as I have already written, with a finite homogeneous Markov chain (they are used, for example, for computer text generation). However, I doubt that it would be possible to predict large outliers well, due to their rarity.

And most importantly - the prediction will be in a probabilistic sense, i.e. with some probability of error.
 
Yuriy Asaulenko:
Think, mathematician, think. Let's try it three times.

Let the horse think. I have no time. I'm galloping towards the native grail.

 
Alexander_K2:

Let the horse think. I have no time. I am galloping towards my home grail.

Not my business, of course, but that is one of the reasons for your failures.
 
Yuriy Asaulenko:
Not that it's any of my business, but that's one of the reasons you're failing.


You should have said so already, instead of competing to see who is smarter.

 
Evgeniy Chumakov:

You should have said so already, instead of competing to see who is smarter.

A man learns only by himself. There is no other way.
(To teach a scientist is only to spoil).
Yes, he does not need it).