Adding new lines to data feed while downsampling on data with price bars

Akul Kanojia

Hi all,
I downsampled data feed using filters to make price bars. Now there is some information that I want to retain with the new data feed for eg, total qty traded within the price bar, which I did on by updating a column names Qty.

But let's say I want to retain additional information eg timeDelta, priceDelta, etc., I would require additional columns in the data itself.

How to add such columns while applying a filter to a data feed?

backtrader

@Akul-Kanojia said in Adding new lines to data feed while downsampling on data with price bars:

How to add such columns while applying a filter to a data feed?

You said you have used filters, but which?

If your filter is custom you will have to copy the data from the input column to the output column.

Akul Kanojia

Yes, I have a custom filter, I'm trying to aggregate data in price bars. In the small illustration below, I'm trying to make a new line 'priceDiff'. The line 'Ltp' is already defined in a custom feed deriving GenericCSVData class.

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Dec 11 06:41:45 2019

@author: akul
"""
#!/usr/bin/env python
# -*- coding: utf-8; py-indent-offset:4 -*-
###############################################################################
#
# Copyright (C) 2015, 2016, 2017 Daniel Rodriguez
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
###############################################################################
from __future__ import (absolute_import, division, print_function,
                        unicode_literals)


from . import Filter


__all__ = ['RangeBarFilter']


class RangeBarFilter(Filter):
    '''Modify the data stream to create rangebars of price change similar to Renko

    Params:

      - ``size`` (default: *None*) The size to consider for each brick

      - ``autosize`` (default: *20.0*) If *size* is *None*, this will be used
        to autocalculate the size of the bricks (simply dividing the current
        price by the given value)

      - ``align`` (default: *1.0*) Factor use to align the price boundaries of
        the bricks. If the price is for example *3563.25* and *align* is
        *10.0*, the resulting aligned price will be *3560*. The calculation:

          - 3563.25 / 10.0 = 356.325
          - round it and remove the decimals -> 356
          - 356 * 10.0 -> 3560

    See:
      - http://stockcharts.com/school/doku.php?id=chart_school:chart_analysis:renko

    '''

    params = (
        ('size', None),
        ('autosize', 250.0),
        ('align', 1.0),
    )

    def nextstart(self, data):
        o = data.lines.Ltp[0]
        o = round(o / self.p.align, 0) * self.p.align  # aligned
        self._size = self.p.size or float(o // self.p.autosize)
        self._top = int(o) + self._size
        self._bot = int(o) - self._size
        self.initialPrice = data.lines.Ltp[0]
        
    def next(self, data):
        
        if (data.lines.Ltp[0] > self._top) or (data.lines.Ltp[0] < self._bot):
            data.lines.priceDiff[0] = data.lines.Ltp[0] - self.initialPrice
            self.initialPrice = None
            o = data.lines.Ltp[0]
            o = round(o / self.p.align, 0) * self.p.align  # aligned
            self._size = self.p.size or float(o // self.p.autosize)
            self._top = int(o) + self._size
            self._bot = int(o) - self._size
            return False  # length of data stream is unaltered

        if(not self.initialPrice):
           self.initialPrice  = data.lines.Ltp[0]
        data.backwards()
        return True  # length of stream was changed, get new bar

The error I get it something like this,

Traceback (most recent call last):
  File "testpandas.py", line 52, in <module>
    cerebro.run()
  File "/home/akul/Documents/backtrader/cerebro.py", line 1127, in run
    runstrat = self.runstrategies(iterstrat)
  File "/home/akul/Documents/backtrader/cerebro.py", line 1212, in runstrategies
    data.preload()
  File "/home/akul/Documents/backtrader/feed.py", line 688, in preload
    while self.load():
  File "/home/akul/Documents/backtrader/feed.py", line 523, in load
    retff = ff(self, *fargs, **fkwargs)
  File "/home/akul/Documents/backtrader/flt.py", line 48, in __call__
    self.next(data)
  File "/home/akul/Documents/backtrader/filters/rangeBarFilter.py", line 94, in next
    data.lines.priceDiff[0] = data.lines.Ltp[0] - self.initialPrice
AttributeError: 'Lines_LineSeries_DataSeries_OHLC_OHLCDateTime_Abst' object has no attribute 'priceDiff'

Akul Kanojia

@backtrader Please look at the above illustration. Thanks in advance.

backtrader

Let me tell you what I have done

1. See what the error is ...

@Akul-Kanojia said in Adding new lines to data feed while downsampling on data with price bars:

File "/home/akul/Documents/backtrader/filters/rangeBarFilter.py", line 94, in next
    data.lines.priceDiff[0] = data.lines.Ltp[0] - self.initialPrice
AttributeError: 'Lines_LineSeries_DataSeries_OHLC_OHLCDateTime_Abst' object has no attribute 'priceDiff'

2. Read what your intention is

@Akul-Kanojia said in Adding new lines to data feed while downsampling on data with price bars:

I'm trying to make a new line 'priceDiff'. The line 'Ltp' is already defined in a custom feed deriving GenericCSVData class.

and I have to read that because the custom data feed is not there (probably the key to the problem)

3. Conclude that you haven't defined a line named priceDiff in your custom data feed. Unfortunately, lines are not created out of thin air.