Machine learning in trading: theory, models, practice and algo-trading - page 2291
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
my research shows the opposite picture
In Figure 2, how is the confidence interval plotted?
In figure 2, how is the confidence interval constructed?
standard akf from pandas package, I do not know how exactly. But you can see that the 1st lags are clearly not in it
in the last article the seasonal ones are confirmed purely through the MO
and the last trades in the account confirm it too
Colleagues,tell me from experience.
I wondered if it makes sense to monitor the weights of the input layer (inputs are normalized) during training? Does it give something realistic to assess the significance of inputs?
I use the library from Dmitriy Gizlykfor experiments.
I know that by unloading the data into R or Python I can calculate all sorts of niche indices. But so far has not reached them, and it is convenient that his solution on the video card is almost "flying".
In general, does it make sense to monitor the weights of inputs for simplicity, or in any case, you must first conduct a detailed analysis of the input data?
Colleagues,tell me from experience.
I wondered if it makes sense to monitor the weights of the input layer (inputs are normalized) during training? Does it give something realistic to assess the significance of inputs?
I use the library from Dmitriy Gizlykfor experiments.
I know that by unloading the data into R or Python I can calculate all sorts of niche indices. But so far have not got to them, and it is convenient that his solution on the video card is almost "flying".
In general, does it make sense to monitor the weights of inputs for simplicity, or in any case you should first conduct a detailed analysis of the input data?
You can assess the impact of signs through the weights
Colleagues,tell me from experience.
I wondered if it makes sense to monitor the weights of the input layer (inputs are normalized) during training? Does it give something realistic to assess the significance of inputs?
I think that no, and even in the process of training, what is the point?
It's more like a developer's trick, or when you know exactly what you're looking for and what you're monitoring it for, and if you don't know you don't need it.
Why I'm writing about "Martin (grid) on MO" - there are almost unlimited possibilities to modify strategies in contrast to the usual discretionary trading. Other trade distributions, other dependencies.
It seems to me that you're moving towards more risk...
You have to move towards entry accuracy, everything else is secondary,
entry accuracy , it's the minimum risk + you always know that the system has stopped working with a minimum loss of money.
If you do not know what's wrong with the grid, you get the maximal risk, and the maximal loss of money.
It seems to me that you are moving towards more risk...
You need to move towards entry accuracy, everything else is secondary,
accuracy of entry, it's the minimum risk + you always know that the system stopped working with a minimum loss of money.
As for the nets, it's the maximal risk + you'll never know what's gone wrong and therefore the maximal loss of money
I'm not going anywhere.
There can be a lot of different nets.
I figured no one's ever done it.
But you can see that the 1st lags are clearly not in it
lag 50 in pandas, about the same number of first counts correlate.
There may be false correlations, that's why I took increments, it's almost analogous to cointegration.
lag 50 in pandas, about the same number of first readings correlate.
There can be false correlations, that's why I took increments, it's almost analogous to cointegration.
There's only noise in increments.
how do you find 24 periodic cycles in 1 increments
Why am I writing about martin (grid) on MO - there are almost unlimited possibilities to modify strategies, in contrast to the usual discretionary trading. Other trade distributions, other dependencies.
Make the second output of the net to calculate the lot. Or use the network's confidence as a lot multiplier.