BackTrader 5 : Indicators

Using Indicators

• Indicators can be used in two places in the platform:
  • Inside Strategies
  • Inside other Indicators
• Indicators in action
  • Indicators are always instantiated during __init__ in the Strategy
  • Indicator values (or values thereof derived) are used/checked during next
• There is an important axiom to take into account:
  • Any Indicator (or value thereof derived) declared during __init__ will be precalculated before next is called.
• During __init__
  • line object 에 접근해서 계산할수 있으며 계산이 끝난 결과값에도 재 계산 가능
  •  
  • sma = bt.SimpleMovingAverage(self.data.close) close_sma_diff = self.data.close - sma
• During next

  • close_over_sma = self.data.close > self.sma

• Indicator Plotting
  
  • operation 으로부터 indicator plot 가능, line plot 불가능

• Controlling plotting
  • indicator 안에 plotinfo 가 저장되어지며 dict 이거나 orderdict 으로 되어진다
  • The value can be later accessed (and set) as follows (if needed)

    • myind = MyIndicator(self.data, someparam=value)
      myind.plotinfo.subplot = True

    •  마지막 plot 할때 보조지표로써 추가용
• Params
  • plot (default: True)
  • subplot (default: True)
  • plotname (default: '') - 이거 은근히 중요, plot 오류 대부분 name 지정 안해서 생기는 오류
  • plotabove (default: False) - plot 커스텀할때 cash flow 와 equity 같은 경우 위가 좋음
  • plotlinelabels (default: False) 
  • plotymargin (default: 0.0) - pixel margin 느낌처럼 axis 간격 margin 설정
  • plotyticks (default: []) - ytick 설정
  • plothlines (default: []) - 수직화 설정?
  • plotyhlines (default: [])

Indicator Development

• custom Indicator
• The following is needed:
  • A class derived from Indicator (either directly or from an already existing subclass)
  • Define the lines it will hold
  • An indicator must at least have 1 line. If deriving from an existing one, the line(s) may have already be defined
  • Optionally define parameters which can alter the behavior
  • Optionally provided/customize some of the elements which enable sensible plotting of the indicators
  • Provide a fully defined operation in __init__ with a binding (assignment) to the line(s) of the indicator or else provide next and (optionally) once methodsBe it not the case, at least a next has to be provided where the indicator must assign a value to the line(s) at index 0
  • Optimization of the calculation for the runonce mode (batch operation) can be achieved by providing a once method.
  • If an indicator can be fully defined with logic/arithmetic operations during initialization and the result is assigned to the line: done

  • class DummyInd(bt.Indicator):
        lines = ('dummyline',)
    
        params = (('value', 5),)
    
        def next(self):
            self.lines.dummyline[0] = max(0.0, self.params.value)
    
        def once(self, start, end):
           dummy_array = self.lines.dummyline.array
    
           for i in xrange(start, end):
               dummy_array[i] = max(0.0, self.params.value)

  •  

Mixing Timeframes in Indicators

Background: Indicators are smart dumb objects.

• They are smart because they can make complex calculations.
• They are dumb because they operate with no knowledge of what sources are providing the data for the calculations

• pivotpoint = btind.PivotPoint(self.data1)
  sellsignal = self.data0.close < pivotpoint.s1

Reference

https://note.com/note_rin_note/n/n0fd69fe04a31?magazine_key=m0b66916df1eb

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