Very cool, thanks!
There is one minor annoyance that could be easily fixed:
The chart frame is always too small and cannot be maximized, you always have to scroll to see the dates. At the same time, the code and log frames leave an empty space unused at the bottom of the screen. Could you move those two down and show the whole chart again?
Some more ideas:
1) copy to clipboard buttons for source code and log would be great (especially for the log, since scrolling increments get too large with very long logs so you cannot read them properly anymore without copying to an external text editor)
2) Is the fee value from the settings tab somehow accesible from within the code? (context.fee for example?)
3) Right now, you always have to start with fiat. Could you add an "initial balance (BTC)" option to the settings tab?
4) as already mentioned by someone else, graphical display of buys and sells (green/red dot) in the chart would be extremely helpful.
Topic: CryptoTrader.org - First Web-based Algorithmic Trading Platform for Bitcoin - page 6. (Read 37851 times)
August 10, 2013, 06:45:11 AM
August 09, 2013, 05:42:42 PM
New Feature
Bookmark manager - allows to view and manage all your bookmarked backtests. Also includes a collection of some generic strategies.
Bookmark manager - allows to view and manage all your bookmarked backtests. Also includes a collection of some generic strategies.
August 09, 2013, 06:03:24 AM
All MtGox historical data since July 2010 was uploaded and now can be used in backtests.
many thanks !
August 07, 2013, 05:14:43 PM
Could there be an API function so we can see if we made a profit on the last sell, and how much we made/lost?
It would be very useful as a function for learning.
The bot I'm working on has multiple algorithms, and I'd like to be able to adjust their influence based on how many times they were correct.
Matthew:out
It would be very useful as a function for learning.
The bot I'm working on has multiple algorithms, and I'd like to be able to adjust their influence based on how many times they were correct.
Matthew:out
I'll look at how this can be implemented. Meanwhile, in this thread https://bitcointalk.org/index.php?topic=261086.20 you can find some examples of algorithms that access internal portfolio object to calculate extended statistics. For example https://cryptotrader.org/backtests/ww9DXh5TPXRcwfhcr
August 07, 2013, 04:56:44 AM
Could there be an API function so we can see if we made a profit on the last sell, and how much we made/lost?
It would be very useful as a function for learning.
The bot I'm working on has multiple algorithms, and I'd like to be able to adjust their influence based on how many times they were correct.
Matthew:out
It would be very useful as a function for learning.
The bot I'm working on has multiple algorithms, and I'd like to be able to adjust their influence based on how many times they were correct.
Matthew:out
August 06, 2013, 03:14:33 PM
Bug? I attempt to select M15 candle and the interface flashes then selected M1 candle for me.
- Create a new backtest
- Select M15 (it stays)
- Attempt to change the time period (say to May 1st)
- M1
- Create a new backtest
- Select M15 (it stays)
- Attempt to change the time period (say to May 1st)
- M1
The latest version should fix this issue
I believe 4 months in 15m intervals will generate too many candles, so I guess this was limited intentionally. I agree it shouldn't switch to 1m though.
Also, when entering www.cryptotrader.org instead of https://cryptotrader.org it redirects you to bitcoinpedia.ru with a 502 error.
Also, when entering www.cryptotrader.org instead of https://cryptotrader.org it redirects you to bitcoinpedia.ru with a 502 error.
Fixed as well, though i'm surprised that someone still uses "www" prefix
Of course, but if a strategy gives excellent results in both 2013, 2012 and 2011, it's far more reliable than having excellent results in 2013 and bad results in 2011-2012. Some luck or a specific configuration could have biased the results
All MtGox historical data since July 2010 was uploaded and now can be used in backtests.
August 06, 2013, 06:39:50 AM
I agree that the more data there is the better strategies you can develop. But keep in mind that markets change, I don't think the way the market behaved in 2011 is comparable to it's current behaviour.
Of course, but if a strategy gives excellent results in both 2013, 2012 and 2011, it's far more reliable than having excellent results in 2013 and bad results in 2011-2012. Some luck or a specific configuration could have biased the results
August 06, 2013, 02:40:25 AM
I believe 4 months in 15m intervals will generate too many candles, so I guess this was limited intentionally. I agree it shouldn't switch to 1m though.
Also, when entering www.cryptotrader.org instead of https://cryptotrader.org it redirects you to bitcoinpedia.ru with a 502 error.
Also, when entering www.cryptotrader.org instead of https://cryptotrader.org it redirects you to bitcoinpedia.ru with a 502 error.
August 05, 2013, 05:11:29 PM
Bug? I attempt to select M15 candle and the interface flashes then selected M1 candle for me.
- Create a new backtest
- Select M15 (it stays)
- Attempt to change the time period (say to May 1st)
- M1
- Create a new backtest
- Select M15 (it stays)
- Attempt to change the time period (say to May 1st)
- M1
August 05, 2013, 01:48:10 PM
Do not do that only for me but if other are interested too, it would be useful, the more data we have, the more reliable a strategy can be, especially if you test on the long-term! In 2011 we had the biggest bitcoin bubble until now followed by a long deflation, it can be interesting to backtest on it, the years 2012 and 2013 are very different
I agree that the more data there is the better strategies you can develop. But keep in mind that markets change, I don't think the way the market behaved in 2011 is comparable to it's current behaviour.
August 05, 2013, 12:42:26 PM
Seems to be working now, must be " instead of '
It should be no difference. I will look into it.
Since few days, I can't backtest on 2011 data 
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
At this time, Cryptotrader does not allow backtesting on historical data for the periods prior to 2011/06/27. Though, if you really need that, i could upload market data until June 2011 to the database.
Do not do that only for me but if other are interested too, it would be useful, the more data we have, the more reliable a strategy can be, especially if you test on the long-term! In 2011 we had the biggest bitcoin bubble until now followed by a long deflation, it can be interesting to backtest on it, the years 2012 and 2013 are very different
August 05, 2013, 06:37:01 AM
Seems to be working now, must be " instead of '
It should be no difference. I will look into it.
Since few days, I can't backtest on 2011 data
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
At this time, Cryptotrader does not allow backtesting on historical data for the periods prior to 2011/06/27. Though, if you really need that, i could upload market data until June 2011 to the database.
August 04, 2013, 07:12:31 PM
Since few days, I can't backtest on 2011 data
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
More precisely, I can't go before 2011/06/27
Is there some bad data in the database ?
August 03, 2013, 06:29:00 PM
Seems to be working now, must be " instead of '
August 02, 2013, 06:08:57 PM
In case of error, sendEmail logs an error message. if there aren't any error, then possible reasons can be:
- at high trading frequency candles are not generated continuously due to low market volatility, so the code may not get called as frequently as you expect
- email messages are filtered by Spam filters.
You can add some debug output to the handle method or provide a link to backtest you ran here or via PM, so i can take a look.
- at high trading frequency candles are not generated continuously due to low market volatility, so the code may not get called as frequently as you expect
- email messages are filtered by Spam filters.
You can add some debug output to the handle method or provide a link to backtest you ran here or via PM, so i can take a look.
August 02, 2013, 05:50:41 PM
Code:
sendEmail 'xxx'
[/quote]
I added this to my code to test it, but I don't get any emails. Is there a simple way to test it?
[/quote]
I added this to my code to test it, but I don't get any emails. Is there a simple way to test it?
August 02, 2013, 07:10:36 AM
Good idea IMO. Added it to the roadmap.
good news !
August 02, 2013, 06:47:33 AM
Good idea IMO. Added it to the roadmap.
August 01, 2013, 03:10:56 PM
A simple idea (but I don't know if it's easy to make) : it could be great to have graphical representation of the BUY and SELL entry points, like this :
Would be a quick way to see if a strategy works
Would be a quick way to see if a strategy works
July 31, 2013, 01:43:04 PM
This is great!
Thanks
How do I set the name for a Trader?
Can you add something similar for Backtests as well, so one can have a collection of trading strategies? Right now bookmarking everything manually is kind of tedious ..
Thanks
How do I set the name for a Trader?
Can you add something similar for Backtests as well, so one can have a collection of trading strategies? Right now bookmarking everything manually is kind of tedious ..
Thanks for feedback! The features you are asking for will be added in the next release.
Jump to: