Discussion Forum - QuantConnect.com

Gene Wildhart INVESTOR

| Reward Discussion

I have a need to keep a history (rolling window) of X number of past indicator values. In the simple code below, I'm trying to keep the past 180 fifteen minute periods of a bollinger band by storing them in a RollingWindow just as I do with the consolidated TradeBars. Unfortunately, it is not working and all elements of the the _bbwindow data structure end up being the same, presumably because upon calling the _bbwindow.Add() function to store the bollinger band, it is probably storing a pointer to the bollinger band object. Can someone help explain what the preferred method of storing historical values of indicators are?

Thanks so much!!

using System;

using System.Collections.Generic;

using QuantConnect.Data.Consolidators;

using QuantConnect.Indicators;

using QuantConnect.Data.Market;





namespace QuantConnect.Algorithm 

{   

	public class ForexBollinger : QCAlgorithm

	{

		BollingerBands _bb;



		//Use our new consolidator class - 15 minutes / 15 bars joined.

		decimal _price;

		decimal stopPrice;

		string symbol = "EURUSD";

		static int Hist = 180;

		static int Inputs = 4;

		RollingWindow _datawindow = new RollingWindow(Hist);

		RollingWindow _bbwindow = new RollingWindow(Hist);



		public override void Initialize()

		{

			SetStartDate(2014, 5, 1);         

			SetEndDate(2015,6,4); 

			SetCash(200000);

			AddSecurity(SecurityType.Forex, symbol, Resolution.Minute);

			_bb = new BollingerBands(20, 2, MovingAverageType.Simple);

	

			var fifteenConsolidator = new TradeBarConsolidator(TimeSpan.FromMinutes(15));

			fifteenConsolidator.DataConsolidated += OnDataFifteen;

			SubscriptionManager.AddConsolidator(symbol,fifteenConsolidator);

			RegisterIndicator(symbol, _bb, fifteenConsolidator, x => x.Value);

		}



		public void OnData(TradeBars data) 

		{   

			if (!_bb.IsReady) return;



			if (!Portfolio.HoldStock) 

			{

				Order("EURUSD", 1000);

				Debug("Purchased EURUSD on " + Time.ToShortDateString());

			}



			foreach(string symbol in Securities.Keys){

				if (Securities [symbol].Holdings.IsLong) {

					if (data[symbol].Close <= stopPrice) {

						Liquidate(symbol);

						Debug ("Hit StopLoss: " + data[symbol].Close);

					}

				}

				if (Securities [symbol].Holdings.IsShort) {

					if (data[symbol].Close >= stopPrice) {

						Liquidate(symbol);

						Debug ("Hit StopLoss: " + data[symbol].Close);

					}

				}

			}



		}



		private void OnDataFifteen(object sender,TradeBar consolidated)

		{

			_price = consolidated.Close;

			if (!_bb.IsReady) return;

			_datawindow.Add(consolidated);

 // **** _bbwindow does not really keep a history of the Bollingerbands!!

			_bbwindow.Add (_bb);

			if (!_datawindow.IsReady) return;

			if (!_bbwindow.IsReady) return;



			Plot("BB", "Price", _price);

			Plot("BB", _bb.UpperBand, _bb.MiddleBand, _bb.LowerBand);   

		}



		// Fire plotting events once per day:

		public override void OnEndOfDay() 

		{

			//Log("EndOfDay");

		}



	}

}

Update Backtest

Gene Wildhart

Accepted Answer

Update Backtest

Notebook

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.

Jared Broad

STAFF ,

You're almost perfect @Gene, the only change was to make the RollingWindow store decimal values. All indicators have operator overloads so strictly speaking they are classes, but you can use them like numbers - like this:

decimal x = _bb + _bb2;

RollingWindow _bbwindow = new RollingWindow(Hist);

The by making the rolling window use decimal type, you're triggering the operator overload to return a decimal value for the BB value which will be stored permanently as a value-type (instead of reference type like a class).

The UpperBand/LowerBand properties are sub-indicators, so they have the same class-operator-overload behavior. You should probably save these to the rolling window, I think the middle BB band is just the SMA.

Michael Handschuh

50.5k ,

Hey @Gene!

I've attach a sample that may be useful if you wanted to save off the entire state of the BollingerBands object. I've defined a type to hold the standard deviation and the upper,middle,lower bands of the indicator. I hope this helps!

Gene Wildhart

11.5k ,

Thanks Michael! Very useful indeed, and much cleaner than saving each individual band member to its own rolling window.

Dan Hill

529 ,

Hey @MichaelH

Has something changed with the method?

I noticed that this example now thows an error:

Backtest Error: Error initializing algorithm: This may be because history is using fake data while pre-analyzing an algorithm for a backtest

Backtest Error: Error initializing algorithm: Type mismatch found between consolidator and symbol. Symbol: EURUSD expects type QuoteBar but tried to register consolidator with input type TradeBar

Alexandre Catarino

100.3k ,

Since a forex bar is now QuoteBar (they include Bid/Ask information).
We need to update the code above accordingly:

// Declare RollingWindow of type QuoteBar instead of TradeBar

RollingWindow<QuoteBar> _datawindow =

    new RollingWindow<QuoteBar>(Hist);

In Initialize, we change the consolidator:

var fifteenConsolidator = 

    new QuoteBarConsolidator(TimeSpan.FromMinutes(15));

Finally, we need to change the type of the event handler:

private void OnDataFifteen(object sender, QuoteBar consolidated)

{

    //

}

Please checkout the attached project with all the changes.

Gene Wildhart INVESTOR

Update Backtest

Notebook

To unlock posting to the community forums please complete at least 30% of Boot Camp.
You can continue your Boot Camp training progress from the terminal. We hope to see you in the community soon!

Platform

QuantConnect Community

Alpha League Competition: $1,000 Weekly Prize Pool

Filter Discussion by Tags

240100 Quants.

Online Now

Stay Connected

Preferred method of keeping a history (RollingWindow) of Indicators

Started By:

Join QuantConnect for Free Sign Up

Platform

SIGN IN

QuantConnect Community

Alpha League Competition: $1,000 Weekly Prize Pool

Filter Discussion by Tags

240100 Quants.

Online Now

Stay Connected

Preferred method of keeping a history (RollingWindow) of Indicators

Started By:

Join QuantConnect for Free Sign Up