@jinwook-jang I think I have to use "dynamic indictor". https://www.backtrader.com/blog/posts/2018-02-06-dynamic-indicator/dynamic-indicator/

@hfrog713 said in Correct way to use nested If statements on a SignalStrategy:

delta_under_flag = bt.If(delta_below_zero, 1, 0) hl2_under_flag = bt.If(delta_and_hl2_down, 1, 0)

Seems to be working fine, i accidentally had 1 on both the buy and the sell signal.
Now everything seems to be working well.
Please let me know if there are any comments.

@run-out said in indicator on resampled data:

self.zz = ZigZag(self.datas[1], period=2, retrace=0.05, minbars=2)

Hi! I've done everything as you had adviced and i have data for datas[1], but nothing is calculated for datas[0]. COuld you advice something to solve it?

def __init__(self): # Keep a reference to the "close" line in the data[0] dataseries self.dataclose1 = self.datas[0].close self.dataclose30 = self.datas[1].close # To keep track of pending orders self.order = None self.zigzag1 = ZigzagNiko.ZigZag(self.datas[0], period=2, retrace=0.05, minbars=2) self.zigzag30 = ZigzagNiko.ZigZag(self.datas[1], period=2, retrace=0.05, minbars=2) self.ZZLevels1 = pd.DataFrame(columns = ['ZZ1']) self.ZZLevels30 = pd.DataFrame(columns = ['ZZ30'])

and second part when i check data is:

def next(self): # Simply log the closing price of the series from the reference self.log('Close1, %.2f' % self.dataclose1[0]) self.log('Close30, %.2f' % self.dataclose30[0]) if self.zigzag30.lines.zigzag[0]>0 or self.zigzag1.lines.zigzag[0]>0: self.log('ZZ30, %.2f' % self.zigzag30.lines.zigzag[0]) self.log('ZZ10, %.2f' % self.zigzag1.lines.zigzag[0]) # Check if an order is pending ... if yes, we cannot send a 2nd one if self.order: return df1 = pd.DataFrame({'ZZ1': [self.zigzag1.lines.zigzag[0]]}) df30 = pd.DataFrame({'ZZ30': [self.zigzag30.lines.zigzag[0]]}) self.ZZLevels30=self.ZZLevels30.append(df30, ignore_index=True) self.ZZLevels30=self.ZZLevels30.dropna() self.ZZLevels1=self.ZZLevels1.append(df1, ignore_index=True) self.ZZLevels1=self.ZZLevels1.dropna() print(self.ZZLevels30) print(self.ZZLevels1)

and finally i add data this way:

data = exporter.download(asset_id, market=Market.FUTURES, timeframe=Timeframe.MINUTES1, start_date=fromdate1) data['<DATE>'] = data['<DATE>'].apply(lambda x: str(x)) data['<TIME>'] = data['<TIME>'].apply(lambda x: str(x)) data['<DATE>'] =data['<DATE>']+' '+ data['<TIME>'] data['<DATE>'] = data['<DATE>'].apply(lambda x: pd.to_datetime(str(x), format='%Y-%m-%d %H:%M:%S.%f')) rts = pd.DataFrame(data) rts = rts.set_index('<DATE>') rts= rts.loc[:,['<OPEN>','<HIGH>', '<LOW>', '<CLOSE>','<VOL>']] rts.index.names = ['Date'] rts.columns=['Open','High','Low','Close','Volume'] #print(rts) data1 = bt.feeds.PandasData(dataname=rts) cerebro.adddata(data1,name = '1M') # Add the Data Feed to Cerebro data30 = cerebro.resampledata(data1, name='data30', timeframe=bt.TimeFrame.Minutes, compression = 30) # Add the Data Feed to Cerebro cerebro.adddata(data30,name = '30M')

and finally i have data for ZZ30 and nothing for ZZ1:
9b12a815-f578-42f5-a65e-e18776e8fdc4-image.png

@asus-rog Have you tried to use it for 2 timeframes? I've tried with resampling in backtrader, resampling in pandas, but nothing helps. Everytime I've got just Nan.

Hi, Nikolai!
Could you please advice how to use your indicator for 2 timeframes?

@vladisld perhaps I'm not pretty correct but after loading 2 timeframes i see correct data for my indicator when I debug each step and I see an array with correct calculations but it is zero because when i add second timeframe indicator starts to recalculate every time when next starts and all previous data are Nan. SO my question is how to call proper calculations from my indicator and not allow it to recalculate?
54330bb6-77b0-42ef-baa2-be70b93d10a9-image.png

@run-out You are right, I thought there may be some ready option for it, but just simple:

def __init__(self): self.coins_indicators[d]["pnl"] = 0 def notify_trade(self, trade): if trade.isclosed: self.coins_indicators[trade.data]["pnl"] += trade.pnl

Get's the job done, thank you!

@vladisld Perfect, thanks!

@adarakchian so in both your way and the way I've proposed - the main issue is to replace the array.array with other container.

Using the QBuffer mode seems to be less intrusive way, while in your case the container type is changed forcefully.

In both cases the target seems to be the same: the container type should not make an assumption about the value type used for line data.

One more note: there are some assumptions about the underlying container type inside the LineBuffer - and this is reflected in the LineBuffer.mode ( QBuffer or Unbounded).

Depending on this mode, different methods are used for slicing and other operations inside the LineBuffer methods

Given that, it is important to keep the container type compatible with what LineBuffer logic is expecting IMHO

@lin-ding said in Sizer based on ATR error I need help:

cerebro.addsizer(AtrBasedPositionSize)

Please take a look at the sizers documentation, there is an example of how to define the sizer with custom parameters.

https://www.backtrader.com/docu/sizers/sizers/

On a different matter:

Please use the backtick or grave accent to wrap you code - this way it will be easier for everybody to read you posts (see the header on this page)

The best way to extract data from your backtest is Analyzers.

You can use built in analyzers or as in your case, build a custom analyzer to track the data during each iteration of the back tests.

At the end, you will usually end up with a dictionary (if you build it that way) out of which you can build your dataframe.