Help with starting using BT

Marina Zaitseva

I'm a beginner in BT. First, I tried to create the easiest indicator, using BT. The Code in TradeStation:

I want to get same result for f_LDirNum using BT.

I copied sma.py to fxd_sma_mod.py
Added fxd_sma_mod.py to init.py
Filled fxd_sma_mod.py with several classes

#!/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 MovingAverageBase, Average
from . import (Indicator, Highest, Lowest, If, Accum)

import backtrader as bt
import math

class f_MovingAverageSimple(MovingAverageBase):
    '''
    Non-weighted average of the last n periods

    Formula:
      - movav = Sum(data, period) / period

    See also:
      - http://en.wikipedia.org/wiki/Moving_average#Simple_moving_average
    '''
    alias = ('f_SMA', 'f_SimpleMovingAverage',)
    lines = ('f_sma',)

    def __init__(self):
        # Before super to ensure mixins (right-hand side in subclassing)
        # can see the assignment operation and operate on the line
        self.lines[0] = Average(self.data, period=self.p.period)

        super(f_MovingAverageSimple, self).__init__()
		
class f_LDir(Indicator):

    alias = ('f_LDir',)
    lines = ('f_lDir',)

    def __init__(self):
        l = self.data.low
        l1 = self.data.low(-1)		
        self.lines.f_lDir =  bt.If(l < l1, l-l+1, l-l-1)
        super(f_LDir, self).__init__()


class f_LDirNum(Indicator):

    alias = ('f_LDirNum',)
    lines = ('f_lDirNum',)

    def __init__(self):
        #Sign = lambda x: math.copysign(1, x) # two will work
        l = self.data.low
        l1 = self.data.low(-1)	
				
        lDir =  bt.If(l < l1, l-l+1, l-l-1)

        lDir1 = lDir(-1)
        deltalDir = abs(lDir - lDir1)		
	
        self.lines.f_lDirNum =  bt.If(deltalDir==2, lDir, 
                                    f_lDirNum1+lDir)									
									
        f_lDirNum1 = self.lines.f_lDirNum		
	
        super(f_LDirNum, self).__init__()

In Jupyter Notebook created a file with a next code

#from datetime import datetime

import backtrader as bt
import datetime
import backtrader.feeds as btfeeds

import pandas
import pandas_datareader
import datetime
import pandas_datareader.data as web

# Create a subclass of Strategy to define the indicators and logic
class BasicIndicator(bt.Strategy):
    # list of parameters which are configurable for the strategy
    params = dict(
        pfast=10,  # period for the fast moving average
        pslow=30   # period for the slow moving average
    )

    def __init__(self):
        
        #sma = bt.indicators.SimpleMovingAverage(self.data)
        f_sma = bt.indicators.f_SMA(self.data, period=15)

        f_lDir = bt.indicators.f_LDir(self.data)
        #f_lDirNum = bt.indicators.f_LDirNum(self.data)        

# Create a data feed      

dataframe = pandas.read_csv(
    'GBP_D_MW.txt',
    parse_dates=True,
    index_col=0,
    )


# Pass it to the backtrader datafeed and add it to the cerebro
data = bt.feeds.PandasData(
                            fromdate = datetime.datetime(2018, 12, 1),
                            todate =  datetime.datetime(2019, 12, 31),    
                            dataname=dataframe,
                            nocase=True,
                           )

cerebro = bt.Cerebro()  # create a "Cerebro" engine instance
cerebro.adddata(data)  # Add the data feed
cerebro.addstrategy(BasicIndicator)  # Add the trading strategy
cerebro.run(stdstats=False)  # run it all

cerebro.plot(iplot=False, style='candlestick',   
             #numfigs = 1,              
             volume=False, 
             plotstyle = 2, 
             zup = False, zdown=False )  # and plot it with a single command

Using this code, i get result, which i expect

I've got same result "lDir", as at blue histogram in TradeStation Chart.
But i need to get a red Histogram "f_LDirNum"of TS.
And this is - my Big Problem, where i need a Help.

I can't write right way a code for a class:

class f_LDirNum(Indicator):

If i change a code in Jupyter NB, adding f_lDirNum

    def __init__(self):
        
        #sma = bt.indicators.SimpleMovingAverage(self.data)
        f_sma = bt.indicators.f_SMA(self.data, period=15)

        f_lDir = bt.indicators.f_LDir(self.data)
        f_lDirNum = bt.indicators.f_LDirNum(self.data)

i get an error

I can understand why, looking at my code
Скриншот 26-08-2019 165502.png

But i'm a beginner in Python too, so i don't know where i can initialize f_lDirNum1 to Zero for Example. In TradeStation i can use:

if BarNumber = 1 then f_lDirNum1 = 0

But in BackTrader Guide I could find how to check a barnumber in TimeSeries.
I tried many variants of algorithm "class f_LDirNum(Indicator):", but couldn't get a Red Histogram from TradeStation Chart.
After 1 week experiments, i decided to ask your help.

Please, help me to code a class f_LDirNum(Indicator)

snevins

@Marina-Zaitseva said in Help with starting using BT:

from TradeStation Chart

Do you mean to have the 1 at the end of f_lDirNum1? You don't have anything defined by that name prior to it executing but you have f_lDirNum. When you run that if statement you are inherently defining f_lDirNum1 so that's why it works when that if statement executes

Marina Zaitseva

@snevins
I'd like to know how to define f_lDirNum1 somehow (0 or 1 or any number) before f_lDirNum will be counted. (if its possible). Or maybe other algorithm to count f_lDirNum.

Briefly describe the task:

I have Condition = Low < Low[1] (I can choose any other condition)

Each time when Condition = True - lDir=1
If Condiiton = True 2 times in а row - lDir=1, f_lDirNum=2
If Condiiton = True 3 times in а row - lDir=1, f_lDirNum=3
If Condiiton = True X times in а row - lDir=1, f_lDirNum=X
.......
Each time when Condition = False lDir=-1
If Condiiton = False 2 times in а row - lDir=-1, f_lDirNum=-2
If Condiiton = False 3 times in а row - lDir=-1, f_lDirNum=-3
If Condiiton = False X times in а row - lDir=-1, f_lDirNum=-X