[Feature Request] Integrate rolling window into indicators

Yuri Lopukhov,

This option already exist 

IndicatorExtensions.Of(Delay(), ...)

If you are satisfied with my answer, please accept it and don't forget to like it. 

Танцуй Россия!

Here is the code snippet from the backtest attached above.

# QC DEMA Delayed

from AlgorithmImports import *

STOCK = 'SPY'; PERIOD = 50; DELAY = 10;

class DemaDelayed(QCAlgorithm):
    def Initialize(self):
        self.SetStartDate(2020, 1, 31)
        self.SetEndDate(2021, 10, 20) 
        self.SetWarmUp(5*PERIOD + DELAY, Resolution.Daily)
        self.stock = self.AddEquity(STOCK, Resolution.Daily).Symbol
        self.dema = self.DEMA(self.stock, PERIOD, Resolution.Daily)
        self.dema_delayed = IndicatorExtensions.Of(Delay(DELAY), self.dema)

    def OnData(self, data):
        if self.IsWarmingUp or not (self.dema.IsReady and self.dema_delayed.IsReady): return  

        self.Plot("Indicator","DEMA", self.dema.Current.Value)
        self.Plot("Indicator", "DEMA_delayed", self.dema_delayed.Current.Value)

Good to know, but this is not the same, what if I need access to any or all of the last 10 values?

Also, my proposal feels more natural to use.

Yuri Lopukhov,

To access any or all of the last N values you may use talib indicators.

If you are satisfied with my answer, please accept it and don't forget to like it.

Here is the code snippet from the backtest attached above

# TaLib DEMA Delayed

from AlgorithmImports import *
import numpy as np
import talib 

STOCK = 'SPY'; PERIOD = 50; DELAY = 10;

class TaLibDEMA(QCAlgorithm):

    def Initialize(self):
        self.SetStartDate(2020, 1, 31)
        self.SetEndDate(2021, 10, 20) 
        self.SetWarmUp(5*PERIOD, Resolution.Daily)
        self.stock = self.AddEquity(STOCK, Resolution.Daily).Symbol
        self.rollingWindow = RollingWindow[TradeBar](5*PERIOD)
        self.Consolidate(self.stock, Resolution.Daily, self.CustomBarHandler)
        
        
    def CustomBarHandler(self, bar):
        self.rollingWindow.Add(bar)         
        if self.IsWarmingUp: return
        if not self.rollingWindow.IsReady: return
        closes = np.flipud(np.array([self.rollingWindow[i].Close for i in range(5*PERIOD)]))

        talib_dema = talib.DEMA(closes, PERIOD)

        self.Plot("Indicator","talib_dema", talib_dema[-1])
        self.Plot("Indicator", "talib_dema_delayed", talib_dema[-DELAY])
        self.Plot("Indicator", "talib_dema_mean", talib_dema[-DELAY:].mean())

Hi Yuri and Vladimir

RollingWindow has been supporting the use with Indicator all along. Please refer to this doc,

Best
Louis

Louis Szeto.

Thank you.
Here is an example of what you recommended for accessing any of the last N values.

How to access all of them to calculate mean() for example?