Machine learning in trading: theory, models, practice and algo-trading - page 1504
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
How does HMM differ from SOM in principle?
I have read about 100 of 650 pages of Kohonen: Self-Organizing Maps (a terrible read from all related fields, but very little not specifically about SOM), in short, "pure SOM" is the formation of neuron connections on static data, the more data have related features, the stronger the neuron connection
SOM: I was arguing with the author of the article about SOM at the beginning of the week , instead of giving an answer the author tried to play a hullabaloo, imho SOM is not a neural network in the classical sense of the NS, just the SOM learning process is similar to the learning NS
I have read about 100 pages of Kohonen: Self-Organizing Maps (a terrible read from all related fields, but very little not specifically about SOM), in short, "pure SOM" is the formation of neuron connections on static data, the more data have related features, the stronger neuron connections
SOM : at the beginning of the week I was arguing with the author of the article about SOM, instead of answers the author tried to play some kind of hullabaloo, imho SOM is not a neural network in the classical sense of the NS, just the learning process SOM similar to the learning NS
yes, it seems that here in general it's customary to answer a question with a question or send somewhere ))
just wondering if it looks like a latent Markovian model or not, if clusters are taken as latent states... I'll have to google it at my leisure
Because clustering uses EM algorithm and in learning Markov chains, for example. Catfish is also clustering
Found a great article by Nicholas, I recommend....
https://www.finam.ru/analysis/newsitem391640076C/
Article is strong, but have read the comments and was horrified, in fact the comments are written one imbeciles, no adequate comments + 99% of them are not on the subject, just as often happens here))
Is the teacher treating on finam now?
The article is strong
The article is generally about how the eminent husband masterfully walked on the rake for several years, but never managed to install Metatrader, but built his own bicycle for many years and said, yes primitive all traders with their primitive indicators, this is the way that leads to the Grail:"Chebyshev, Butterworth, Hilbert filters up to 12th order, Halley filters, Forward and inverse Fourier transform, Forward and inverse cosine transform, Berg, Prony, maximum entropy spectral estimation methods, ARSS predictors, polynomial filtering methods. "
But I`ve never learnt that MT can make a TS in 5 minutes)))
As one physicist says - let's have the state! )))
There's a dime-a-dozen list of these teachers, who are decked out with regalia of 10 or 15 inventions, as one physicist here writes: "Give me the state! )))
There is a state, but not for you, you're not ready ... you have another 5-10 years to optimize machismo, and then you'll distinguish the useful from the mere ...
there is a state, but it's not for you, you're not ready yet ... you have another 5-10 years to optimize Schohastics, and then you will begin to distinguish between useful and Mr. ...
Well, you should have written right away that you paid for the link and advertising in the article is zero information
Why rude?
There is also a state, but it's not for you, you're not ready yet... you have another 5-10 years to optimize, and then you'll start to distinguish the useful from the Mr.
Well, you should have written right away that your link is paid and advertising, the article has exactly zero information.
Why are you being so rude?
just two words - face palm
Had to read it.
read blogs
http://www.kamynin.ru/?cat=8&paged=55
Readings for the week