@scottz1 Hi, Thank you. You can find the report here https://ethz.ch/content/dam/ethz/special-interest/mtec/chair-of-entrepreneurial-risks-dam/documents/dissertation/master thesis/Master_Thesis_Glücksmann_13June2019.pdf

@ab_trader Thanks i read all of you give me and I understood enough about what i asked before

It seems the 'fix' worked pretty well on real time (paper) trading. 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:notify_order: Market Buy Partial, price: 14.63, size: 400, cost: 5852.0, comm: 2.0, ref: 7 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits: p1: 0, p2: 3 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 0, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 1, dt: 2020-01-17 15:45:06, bit size:200, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 2, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits done. ... 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:notify_order: Market Buy Partial, price: 14.63, size: 500, cost: 7315.0, comm: 2.5, ref: 7 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits: p1: 3, p2: 4 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 0, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 1, dt: 2020-01-17 15:45:06, bit size:200, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 2, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 3, dt: 2020-01-17 15:45:08, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 4, dt: 2020-01-17 15:45:09, bit size:182, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits done. ... 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:notify_order: Market Buy Completed, price: 14.629999999999999, size: 682, cost: 9977.66, comm: 3.41, ref: 7 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits: p1: 4, p2: 5 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 0, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 1, dt: 2020-01-17 15:45:06, bit size:200, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 2, dt: 2020-01-17 15:45:06, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 3, dt: 2020-01-17 15:45:08, bit size:100, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:idx: 4, dt: 2020-01-17 15:45:09, bit size:182, bit price:14.63 2020-01-17 15:45:00:HMI-STK-SMART-USD:track:order bits done. Now I wonder how such 'fix' would have been tested in normal scenario ( in dev environment). Usually this requires 'component level testing' of IBStore/IBBroker/IBFeed components. However this will in turn require some mockups for interfaces below (celebro) and above the IB implementation (IBPy2) that are not existing today. How are you usually testing the components outside the core ? I couldn't find such tests in the package, but given that this is already a second issue I've seen in IBStore/IBBroker/IBFeed component it is a good idea to add some tests (at least to cover the issues already found) - volunteering to help with it if accepted by maintainer. Thanks Vlad

@boblef In addition, I have already generated API token on my Oanda account and defined it and account id in my local code so you don't see them in the code I provided.

Explanation here: https://community.backtrader.com/topic/98/float-division-by-zero/5

@vladisld Thanks a lot!

you can have a look at pandafeed.py or chainer.py and implements your own datafeeds, it's very simple, especially for backtest, i think it's no more than load data in _load() but if you want live trading, you have to deal with disconnection and reconnection and backfill the data properly besides, 5000+ is not big, I am backtesting with more than 900k datas around 2min

@tianjixuetu yes it is using Oanda v20 API

Fixed by using self.strategy.indicatorName.lines.lineName[0] Should've read the documentation more closely...

I've got it working, here's some boilerplate code: class EndOfMonth(object): def __init__(self, cal): self.cal = cal def __call__(self, d): if self.cal.last_monthday(d): return True return False class AssetAllocation(bt.Strategy): params = ( ('cal', None), ) def __init__(self): self.add_timer( when=bt.Timer.SESSION_START, allow=EndOfMonth(self.p.cal), ) if __name__ == '__main__': # Create a cerebro entity cerebro = bt.Cerebro() cal = bt.TradingCalendar() cerebro.addcalendar(cal) # Add a strategy cerebro.addstrategy(AssetAllocation, cal=cal) Thanks again.

@vbs while optimize encountered different situation according different strategy, some very simple may run seccessfully,some complicated may not run,some unique may just walk and have no error reporting.I don't have so much time and experience to debug this.Now,I debug the analyzers,there are some errors.

@backtrader said in Optimization with multiple feeds: @oakenstream said in Optimization with multiple feeds: I dont exactly see where it is Neither does the rest of us. I removed this optreturn and it worked: cerebro = bt.Cerebro(optreturn=False) @oakenstream said in Optimization with multiple feeds: Hey, is it possible to optimize with multiple feeds? Yes, we can. Ok, thanks!

I think I fixed it. Seems to work with the observers.

I found the answer in another post. When I'm looping through my data, I rename the dataframe columns to fit the BT convention, then I convert that into a data feed and apply the name. I'll share my Norgate Data Code here for future reference. # Obtain a Pandas dataframe for a given security from Norgate Data priceadjust = norgatedata.StockPriceAdjustmentType.TOTALRETURN padding_setting = norgatedata.PaddingType.NONE start_date = '2019-05-01' #end_date = '2011-12-31' timeseriesformat = 'pandas-dataframe' universe = ['AAPL','TSLA']#,'GE','AMD','SPY'] for ticker in universe: pricedata_dataframe = norgatedata.price_timeseries( ticker, stock_price_adjustment_setting = priceadjust, padding_setting = padding_setting, start_date = start_date, #end_date = end_date, format=timeseriesformat, ) # Rename columns to suit Backtrader pricedata_dataframe.rename( columns={ 'Open':'open', 'High':'high', 'Low':'low', 'Close':'close', 'Volume':'volume', 'Open Interest':'open interest'}, inplace=True) data = bt.feeds.PandasData(dataname=pricedata_dataframe) data._name = ticker cerebro.adddata(data) Anyways, thanks for creating and sharing such an amazing project.

when I write yield in my next() function the strategy doesn't even get to next function

clearner code class CustomIndicator(bt.Indicator): lines = ("custom_line", ) # Price flip is 6 bars params = (('period', 6),) def __init__(self): self.addminperiod(self.params.period) self.line.custom_line =[0.0] * self.data.close.buflen() # => filled with 'nan's when accessed in a strategy