Machine learning in trading: theory, models, practice and algo-trading - page 968
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
In programming, very often people choose what is more convenient rather than what is useful, functional, but watered down to make an extremely questionable choice - do not.
What a convenience, here's Rattle - again, refused to read a file for the test model, while funny that reads the one on which the model was built (if you load it separately), but does not read a similar - and you can not understand what he had a problem.
Well, I'm still stuck with multithreading - not enough memory catastrophically. I found an article about how to convert something there, but I still don't understand how to use it....
SanSanych, the only source of the distribution there is the python site :)
For statistics and machine learning extension IPython and anaconda. Go to the Russian community of opendatascience or watch the videos from Yandex. They have never heard of R at all. So what should be considered a standard? Try python to make your own opinion and compare. Plus knowing python, as you said, will allow you to do not only statistics, but other things, if necessary.
There are other, more reputable sources for language usage statistics. I regularly cite information from those sources.
I have several newsfeeds on statistics - no python anywhere, so no incentive to study it. R texts, on the other hand, are regular.
There's a misconception of me here on the site as an R fan. I am not a fan of any programming language and R in particular, for me a programming language is a tool. But I'm very interested in organizing a specialized, traders' hangout on statistics on this site, and from this interest comes R as a "system of graphs and statistics". And I spend my time in this site for statistics and R is an absolutely accurate expression of my thoughts in this area and no more.
There are other, more authoritative sources for language usage statistics. I regularly cite information from these sources.
I have several newsfeeds on statistics - no python anywhere, so there is no incentive to study it. R texts, on the other hand, are regular.
There's a misconception of me here on the site as an R fan. I am not a fan of any programming language and R in particular, for me a programming language is a tool. But I am very interested in the organization of specialized, traders' hangout on statistics on this site, and from this interest comes R as a "system of graphs and statistics". And I spend my time on this site for statistics, and R is an absolutely accurate expression of thoughts in this area, no more.
No, there is no opinion.
I just write what I liked more, because I'm not a programmer and just hilarious for myself. That's why I'm saying that I liked it.
And would like a connector similar to yours, mb also will have to order if I will continue. And in fact for the time being I don't really care because non-stationarity cannot be killed by statistical methods, as we all found out, but only with God's Providence and highly specialized strategies like arbitrage or night trading.
What a convenience, here's Rattle - again, refused to read a file for the test model, while funny thing is that it reads the one on which the model was built (if you load it separately), and does not read a similar - and you can not understand what he had a problem.
Well, I'm still stuck with multithreading - not enough memory catastrophically. I found an article about how to convert something there, but I still don't understand how to use it....
Just the last months I've been using rattle - it's extremely convenient to check my thoughts and no problems at all. It's more convenient to write a script in R on initial preparation of predictors, save it into .RData, and then load that .RData file into rattle.
Multithreading is from here. You can load all cores and neighboring computers as well.
PS.
A word of advice about learning English. It is ridiculously easy to learn, on the basis of self-discipline and a basic knowledge of grammar.
0. Prepare pieces of paper about 4 * 5 cm
1. take a paragraph of any text and translate it. Each new word you write out on a separate piece of paper, on one side in English, on the other in Russian.
2. Several times a day look through these pieces of paper from both sides: once from the English side, the other from the Russian side.
3. You have to do this regularly.
4. After a couple of weeks you will memorize up to 50 words a day.
5. All you need is a couple thousand words to read English fluently.
In a couple of months you won't have any problems with English, and the problems with the meaning of words, whether in Russian or in English, will come to the fore.
Because non-stationarity is not killed by statistical methods, as we all have figured out, but only by divine providence and highly specialized strategies like arbitrage or night trading.
If it's about non-stationarity, a huge number of publications, the mainstream is GARCH. It is everywhere from high-frequency trading to daily trading.
If about unsteadiness, a huge number of publications, the mainstream is GARCH. Everywhere, from high-frequency to daytime.
I've seen the articles, I haven't seen the charts from the real :)
I've seen the articles, I haven't seen the charts from the real :)
Well, what are you.... I've seen an article once on the GARCH variant of all stocks of the S&P500.
I think that the ideal TS is GARCH+MO. GARCH is especially interesting - the model covers gaps.
Once again I was convinced that R is not my thing :) syntax is almost not highlighted, the code is unreadable, errors are almost not highlighted. The code itself and the language are not aesthetically pleasing
here could be your counterarguments
Yes, you can train an algorithm in 3 lines instead of 5 in python, that's all. The readability in python would be better. I don't see any advantage with MO packages, it's all the same.
Show me an example. I have this in Rstudio/ Everything is hooked up and configurable to user preferences
And errors are instantly shown.
I've asked more than once, write modestly and don't evaluate what you don't know how to use.
Learn the basics.
Good Luck
Show me an example. I have this in Rstudio/ Everything is hooked up and adjusted to the user's preferences
And errors are instantly shown.
I've repeatedly asked, write modestly and do not evaluate what you do not know how to use.
Learn the basics.
Good luck
It's useless to argue, for my taste this editor looks ugly in any color scheme, language too
I think, you've seen the code in python, IDE vscode and jupiter notebook
SanSanych, the only source of the distribution there is the python site :)
For statistics and machine learning extension IPython and anaconda. Go to the Russian community of opendatascience or watch the videos from Yandex. They have never heard of R at all. So what should be considered a standard? Try python to form your own opinion and compare. Plus knowing python, as you said, will allow you to do not only statistics, but other things, if necessary.
It's also an interpreted language, but perfectly highlighted and checks syntax on the fly, not just after running the script, + code folding, notepads and a bunch of other goodies
Stupidity upon stupidity.
1. Packages/modules in Python can be loaded with pip install / conda install and the same name packages may not have the same content at all. And this perl " the only source of the distribution there is the python site :)" should be put in the anals.
2. Neither IPython nor anaconda are extensions for MO. The former is a simple editor with line/block code while anaconda is a package management system (like a repository) and not only Python but also R.
If you want to show off your knowledge, have it. Otherwise you are showing your amateurism.
Be modest.