Machine learning in trading: theory, models, practice and algo-trading - page 1137
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
By exits we mean the marks (any marks), both to buy and to sell. I don't consider any special algorithms of exiting positions, because a sell signal is a natural signal to close a buy order(though it may be wrong, I haven't thought about it)
so this is an interesting task - how to sample marks randomly, from some distribution, or by enumerating distributions in order to obtain a set of unique strategies and choose the best then
in other words, how to change the redistribution function in an efficient way
I don't know, I think it's easier since I've noticed that it's not enough accuracy or signals, so I'm going to do multicurrency, multitime hubs, with separate accounting of orders and positions, like League TS in the next thread.
P.S. Of course the sense in my idea with batch learning may disappear in the process of development, if real real real-time RL adaptive system appears in the meantime:)
Confused by your statistics slightly, the profit 7175 drawdown 2386 and Sharp Ratio 0.1
Your statistics is a bit confusing, the profit is 7,175, the drawdown is 2,386 and the Sharpe Ratio is 0.1
And what should be the Sharpe Ratio? Data from the terminal. Drawdown - I take the maximal value when comparing the drawdown by equity and drawdown by balance. Maybe there was a huge spike which was not closed in time and a succession of losing trades caused by flat.
What should the Sharpe Ratio be? Data from the terminal. Drawdown - I take the maximum value when comparing the drawdown of funds and drawdown on the balance, perhaps there was a huge spike, which was not closed in time and there went a series of losing trades because of the flat.
The ratio profit (loss) \max-slippage should normally be close to the Sharpe coefficient, i.e. in your case ~3(7175/2386) and not 0.1, there are discrepancies of course, but if more than one order, then clearly something is wrong.
In the normal ratio of profit (loss)\max-slippage should be close to the Sharpe coefficient, that is, in your case ~ 3(7175/2386) and not 0.1, there are of course differences but if more than an order, then clearly something is wrong.
From the help the Sharpe coefficient counts differently
"
"
What is the "risk-free rate" - that's the question...
Also, I work on Si futures - maybe that's it? I always have the ratio in tenths.
What is the "risk-free rate" is the question...
It is a conditional risk-free return - % of a bank deposit (in the corresponding currency) in a first-class bank or % of long-term bonds of the U.S. Treasury
This is a notional risk-free return - % of a bank deposit (in the appropriate currency) at a first-class bank or % of long-term U.S. Treasury bonds
The question is where the data is taken by the terminal...
From the reference, the Sharpe ratio is considered differently
"
"
What is the "risk-free rate" - that's the question...
Also, I work on Si futures - maybe that's it? I always have the ratio in tenths.
If I were you, I'd think about what the numbers you're looking at really mean.
Terminals are written by ordinary people who may, for example, too much in the evening, as everyone sometimes, and then Sharp Ratio only in tenths... You should have at least a rough idea of how much Sharp Ratio will have on the chart, this is the most important factor of strategy quality.
PS: Usually the risk-free rate is not taken into account in terminals.
If I were you, I would think about what the real meaning of the numbers you are looking at.
Terminals are written by ordinary people, who may, for example, too much in the evening, as everyone sometimes, and then Sharpe Ratio only in tenths... You should have at least a rough idea of how much Sharp Ratio will have on the chart, this is the most important factor of strategy quality.
PS: Usually in terminals the risk free rate is not taken into account at all.
From the description of the formula is not clear how to check it, for example how to calculate the"arithmetic mean profit for the time of holding a position"?
Maybe it is a matter of a small mathematical expectation?
In any case, I noticed that the higher the indicator, the better, and this is not a little, and the main report is written to the file with the indicators that I understand.
From the description of the formula, it is not clear how to check it, for example, how to calculatethe "arithmetic mean profit for the time of holding the position"?
Maybe it is a small mathematical expectation?
In any case, I noticed that the higher the indicator, the better, and this is not insignificant, while the main report is written to the file with the indicators that I understand.
For my observations I haven't managed to raise sharpen more than 1. And I haven't seen other people's accounts or charts with higher values. Although I may be wrong.