General Code/Help

• S

  problem with live trading(delayed data)
  • Sasha

  10
  0
  Votes
  10
  Posts
  347
  Views

  B

  Your paper trading account uses the same data permissions as your standard account. It won't simply use your real money. You still need permissions. If you have an Interactive Brokers account you should familiarize yourself with what the DOs and DONTs are. As an alternative you can use the faux data demo, but it won't be any good to test any algorithm because the data is completely fake (even in the demo, you cannot use each and every ticker. The forex tickers tend to work most of the time)
• A

  Flipping to short from a long position
  • Agarp

  4
  0
  Votes
  4
  Posts
  168
  Views

  A

  @backtrader Ok, cheers. Thanks for the reply!
• C

  self.array[self.idx + ago]: IndexError: array index out of range - generated from btind.OLS_TransformationN(X, Y, period=self.p.period)
  • Capn_Hacky

  1
  0
  Votes
  1
  Posts
  119
  Views

  No one has replied

• C

  AttributeError: 'str' object has no attribute '_name' - The error is clear - but for the life of me I can't see why it's happening
  • Capn_Hacky

  5
  0
  Votes
  5
  Posts
  260
  Views

  C

  Thank you for taking the time to answer my question; unfortunately the proposed answer is incorrect. This makes sense if you consider that a custom getter function devised in the manner above shouldn't functionally alter the rest of the code. Specfically I've implimented the suggested change from: p0_data = get_data_by_name(pair[0]) p1_data = get_data_by_name(pair[1]) which uses my custom getter function and changed this to p0_data = self.getdatabyname(pair[0]) p1_data = self.getdatabyname(pair[1]) This alteration resulted in an identical error, so indeed is unlikely to have functionally altered anything.
• A

  GenericCSVData
  • augustus

  3
  0
  Votes
  3
  Posts
  226
  Views

  A

  @backtrader Hi Thank you very much for the tip. It turned out to be fruitful.
• A

  [SOLVED] Version: 1.9.73.123 - Still issues with resampling minutes to days?
  • AB

  3
  0
  Votes
  3
  Posts
  177
  Views

  A

  Ok, I solved it... Of course a user error (and I'm not expecting cellular osmosis, LOL). Short story (Fix): When u start dealing with Multiple timeframes and resampling, labeling the timeframe of your datafeed becomes important. It might be important before as well, I just didn't notice. data = bt.feeds.PandasData(dataname=df,timeframe=bt.TimeFrame.Minutes) Looong story (since I typed it up already and explains the story of identifying this issue): Where there? As in version 1.9.66.122 1.9.67.122 Fix compression only scenarios when resampling and resampling after changes in 1.9.66.122 Final correction for rollover fix introduced in 1.9.66.122 Cover use case for mininum period calculation when all operations/indicators don't use the data feeds directly but lines of it Script I'm running: from __future__ import (absolute_import, division, print_function, unicode_literals) import argparse import backtrader as bt import backtrader.feeds as btfeeds import numpy as np def runstrat(): args = parse_args() # Create a cerebro entity cerebro = bt.Cerebro(stdstats=False) # Add a strategy cerebro.addstrategy(bt.Strategy) # Handy dictionary for the argument timeframe conversion tframes = dict( minute=bt.TimeFrame.Minutes, daily=bt.TimeFrame.Days, weekly=bt.TimeFrame.Weeks, monthly=bt.TimeFrame.Months) # Load the Data dataframe = np.load('dataframe_1S_min.npy').item() for symbol, df in dataframe.items(): print(df) data = bt.feeds.PandasData(dataname=df) # Add the Data Feed to Cerebro cerebro.adddata(data) cerebro.resampledata(data, timeframe=tframes[args.timeframe], compression=args.compression) # Run over everything cerebro.run() # Plot the result cerebro.plot(style='bar') def parse_args(): parser = argparse.ArgumentParser( description='Pandas test script') parser.add_argument('--timeframe', default='daily', required=False, choices=['minute','daily', 'weekly', 'monhtly'], help='Timeframe to resample to') parser.add_argument('--compression', default=1, required=False, type=int, help='Compress n bars into 1') return parser.parse_args() if __name__ == '__main__': runstrat() Data screen grab: open high low close volume openinterest 2019-05-02 09:30:00-04:00 184.5000 184.5700 184.5000 184.5100 752 1 2019-05-02 09:31:00-04:00 184.3600 184.7000 183.5600 183.7500 174074 1 2019-05-02 09:32:00-04:00 183.7500 183.8100 183.3800 183.5000 35592 1 ... ... ... ... ... ... ... 2019-05-10 15:57:00-04:00 175.6800 175.7900 175.5400 175.7900 97358 1 2019-05-10 15:58:00-04:00 175.7850 175.9800 175.6400 175.7400 111693 1 2019-05-10 15:59:00-04:00 175.7600 176.0400 175.4700 175.8400 204595 1 [2728 rows x 6 columns] Script instructions: python test_resampling.py --timeframe weekly --compression 1 Result: Works Script instructions: python test_resampling.py --timeframe minute --compression 15 Result: Works Script instructions: python test_resampling.py --timeframe daily --compression 1 Result: Error Traceback (most recent call last): File "test_resampling.py", line 62, in <module> runstrat() File "test_resampling.py", line 41, in runstrat cerebro.run() File "C:\Users\A\AppData\Local\Programs\Python\Python36\lib\site-packages\backtrader\cerebro.py", line 1127, in run runstrat = self.runstrategies(iterstrat) File "C:\Users\A\AppData\Local\Programs\Python\Python36\lib\site-packages\backtrader\cerebro.py", line 1298, in runstrategies self._runnext(runstrats) File "C:\Users\A\AppData\Local\Programs\Python\Python36\lib\site-packages\backtrader\cerebro.py", line 1557, in _runnext dt0 = min((d for i, d in enumerate(dts) ValueError: min() arg is an empty sequence The clue, which solved the problem: After looking at the plots again. It seems that the first data feed gets labeled as "Days" even though it's provided in minutes.
• 

  Getting AttributeError: 'GenericCSV' object has no attribute 'setenvironment' error
  • Burak Aktas

  3
  0
  Votes
  3
  Posts
  269
  Views

  

  Thanks @backtrader I used GenericCSVData and successfully loaded data into cerebro. I was changing the code a lot due to trying loading data phase that is I the error log and original code was pointing different classes.
• 

  Creating only buy orders but still portfolio value seeing a net increase, why // how is this happening?
  • SkullTech

  4
  0
  Votes
  4
  Posts
  117
  Views

  B

  @skulltech said in Creating only buy orders but still portfolio value seeing a net increase, why // how is this happening?: Am I missing something here? It seems you miss that whatever you buy, does still have a value. Ok, you can ask the bank to send you the stocks in paper format and you can burn them. You will then obviously decrease the value of the assets you bought down to zero. In any other case, the actual value will be determined by the last price you have to do the calculations. Your information about at which price you buy the assets and what the last price is: None. But it seems obvious for anyone that the final price is higher.
• 

  Iterate between __init__ and next ?
  • Akash Saxena

  5
  0
  Votes
  5
  Posts
  196
  Views

  

  thanks for this idea ! yes it can be done, but it takes more time and effort as i have to test my code for the EMAs and other indicators that i use. i finally used a strategy to do the above flowchart for EMA line of a particular period. then used optstrategy to iterate over the EMA periods. Not exactly what i started to do, but it gets my job done. thanks for reaching out with help !
• 

  Getting IndexError: array assignment index out of range in cerebro.run()
  • Burak Aktas

  9
  0
  Votes
  9
  Posts
  424
  Views

  

  @backtrader Thanks! I opened another thread since the error I getting was a different one.
• K

  Tips for better/faster code in my custom indicator - Absolute Strength Histogram
  • kicking_potato88

  3
  0
  Votes
  3
  Posts
  202
  Views

  K

  @backtrader Awesome, thks!
• R

  Would appreciate some help with Connors RSI
  • Rudi

  6
  0
  Votes
  6
  Posts
  256
  Views

  R

  Brilliant @backtrader ! I'll make sure to add indicators once I get the hang of creating them!
• S

  Problem with multiple stocks.
  • Sasha

  5
  0
  Votes
  5
  Posts
  251
  Views

  S

  @backtrader Thanks. You helped me a lot ---------------------------------------- 2016-02-16, BUY CREATE LABU in long , 25.4, SMA: 23.916 2016-02-16, Number of stocks: 1850.0 2016-02-17, BUY EXECUTED, Price: 28.60, Cost: 52910.00, Comm 0.00 2016-02-17, Cash 47090.00, Value 100000.00 ---------------------------------------- 2016-02-17, SELL CREATE LABU in long, 28.60, SMA: 24.13, DIFF: 12.60 2016-02-17, Number of stocks 1850.00 2016-02-18, SELL EXECUTED, Price: 25.68, Profit 47508.00, Comm 0.00 2016-02-18, OPERATION PROFIT, GROSS -5402.00, NET -5402.00, SUM -5402.00 ---------------------------------------- 2016-02-18, BUY CREATE LABU in long , 25.68, SMA: 24.016 2016-02-18, Number of stocks: 1731.0 2016-02-19, BUY EXECUTED, Price: 26.96, Cost: 46667.76, Comm 0.00 2016-02-19, Cash 47930.24, Value 94598.00 ---------------------------------------- 2016-02-19, BUY CREATE TQQQ in long , 13.8, SMA: 12.718 2016-02-19, Number of stocks: 1632.0 2016-02-22, BUY EXECUTED, Price: 14.45, Cost: 23582.40, Comm 0.00 2016-02-22, Cash 24347.84, Value 96052.04 2016-02-22, Cash 24347.84, Value 96052.04 ---------------------------------------- 2016-02-22, SELL CREATE LABU in long, 27.80, SMA: 24.14, DIFF: 8.26 2016-02-22, Number of stocks 1632.00 2016-02-23, SELL EXECUTED, Price: 24.20, Profit 39494.40, Comm 0.00 2016-02-23, Cash 63842.24, Value 88645.40 ---------------------------------------- 2016-02-23, SELL CREATE LABU in long, 24.20, SMA: 24.47, DIFF: 2320.00 2016-02-23, Number of stocks 1632.00 2016-02-24, SELL EXECUTED, Price: 24.76, Profit 40408.32, Comm 0.00 2016-02-24, OPERATION PROFIT, GROSS -4722.12, NET -4722.12, SUM -10124.12 2016-02-24, Cash 104250.56, Value 89369.96 The problem was with the size of stocks.
• D

  TypeError: error() takes 2 positional arguments but 4 were given
  • dongiulio

  5
  0
  Votes
  5
  Posts
  503
  Views

  

  @dongiulio thanks for coming back and posting the solution, helped me get through the IB integration. cheers
• 

  Multiple timeframes / what is "current" data for larger timeframes?
  • Pierrot

  4
  0
  Votes
  4
  Posts
  169
  Views

  

  Hi, Thanks for your help. Yes I did started playing with some print statements. Things start to be clearer. Thanks for your help! Bests, Pierre
• C

  Unrealised Return and Days Held in Next?
  • cwse

  2
  0
  Votes
  2
  Posts
  147
  Views

  B

  You can get the unrealized profit and loss by using the commission scheme which corresponds to the given asset. As for the length, you can record the length at which it was executed and compare it against the actual length of the data (that will give you natural days, for trading days, you will need to apply a trading calendar)
• 

  Multiple timeframes not with the same start point / possible?
  • Pierrot

  3
  0
  Votes
  3
  Posts
  166
  Views

  B

  @pierrot said in Multiple timeframes not with the same start point / possible?: Will backtrader be able to handle data coming from different csv files that do not share the same start time? Yes. Even if they start in different millenia, centuries, eons ... @pierrot said in Multiple timeframes not with the same start point / possible?: coming from "1H_data.csv" is the one indeed from the 1st bar Without seeing the data the answer seems to be yes. @pierrot said in Multiple timeframes not with the same start point / possible?: coming from "1D_data.csv" is the one from the 2nd bar (and not the 1st) ? What is the 2nd bar? In any case it would seem strange to have a daily bar with a session closing time of 00:00:00. If your second bar is 2019-05-06 00:00:00, you will get this 2nd bar, but that doesn't make it right, because a daily bar does always complete after intraday bars. @pierrot said in Multiple timeframes not with the same start point / possible?: If not, is there a workaround to be able to have this logic? You haven't specified any logic, rather a half-cooked example. The logic is for sure very clear to you, but not for those reading your message.
• S

  Tws connection problem
  • Sasha

  3
  0
  Votes
  3
  Posts
  122
  Views

  S

  @rami thank's, the problem was with port number. Successfully connected with IB Workstation, and I have another problem.For output, I get Server Version: 76 TWS Time at connection:20190506 01:29:55 ALMT ***** DATA NOTIF: DELAYED ***** DATA NOTIF: NOTSUBSCRIBED
• 

  timer support and IB
  • rami

  3
  0
  Votes
  3
  Posts
  197
  Views

  

  @backtrader, have a couple of questions: how to make algo skeleton, that can trade, not on a day close, but on ticker price at a specific time like 15:30? To make resampling with 30min, and wait for a specific time to validate and execute? How I can get a historical daily price this specific time 15:30 (for example for 200 days)? thank you in advance.
• 

  Crowdfunding support of CCXT support in backtrader?
  • Pierrot

  10
  0
  Votes
  10
  Posts
  401
  Views

  

  @backtrader Dear backtrader you haven't answered the question, which was a question related to the comment you made I am sorry. As you will understand from my other recent posts, I am a beginner in backtrader, so I can only be imprecise. Intended code modifications & additions are basically of the same type than the ones highlighted in Dave Vallance's fork https://github.com/Dave-Vallance/bt-ccxt-store Additons: ccxtbroker.py, ccxtstore.py ccxtfeed.py Modifications: listed in README.md Regarding RACI part, yes, you're completely right, this is a topic that needs to be clarified. I fully agree with what you expose, and have actually already encountered once these questions after a submission I made once (fortran scientific code - clearly not python :)) If it can be addressed thanks to a crowdfunding approach, this is great. Thank you for your answer, and hope it can turn in a positive answer. For other readers, it would be nice to know how many would feel like to contribute some funds, so that everybody has an idea of what amount needs be invested in an effort to even overall the expense (no need to clearly state how much you can afford). If we do be many, this will also demonstrate that the preliminary task of quantifying the development work (if crowdfunding is accepted by @backtrader) is not made in vain. @moontrader : thanks to have dropped a word, this is very positive Thanks again @backtrader & @moontrader , Have a good day. Bests, Pierre