• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Is multi-timeframe backtesting proprietary?

Joined
4/28/10
Messages
73
Points
18
Hello quants,

I would like to ask your opinion on something simple. Is multi-timeframe backtesting proprietary? Specifically strategies involving price action and indicators on 60 min and 5 min charts simultaneously over batches of say 5 years.

Put differently, is there robust software in the market that allows you to test such strategies even to the extent where you can alter the calculations for each indicators. (Simple but infeasible example for illustration - moving average calculations takes every other bar in the look back)

I am pretty such that CQG doesn't allow multi-timeframe backtesting. What about Ninjatrader or CMC Markets?

As to why I'm asking this question ... I want to 1. have a sense of whether already profitable strategies can be improved by fine tuning entries in the smaller time frame 2. validate my role as a quant because I've been using C++ for the past months to developed such a thing.

Thanks,
Donny
 
What do you mean, is it proprietary? There are a number of tools / companies / products available that would allow you to do backtesting, see https://www.quantopian.com/ for an example (you take peoples base algos and modify them or code your own etc, with real historic price data). Or am I misunderstanding your question?
 
Hello Daleholborow,

Yes, I'm aware that there are a range of third party software that does backtesting. That's my question exactly. I am being specific.

What's your take on these software being able to backtest MULTI-TIMEFRAME strategies in ONE pass. This obviously means programming over two sets of data ( i.e. 5 min and 60 min ).

I heard of Quantopian but not sure whether you are given two sets of OHLCs to work with. I know you can't do this with CQG.

Cheers,
Donny
 
If you're stepping through time taking every 5 minute observation, and you wanted to simultaneously check 60 minute observations, whats wrong with simply taking every 12th data point from the first data stream to plug into your second component of your strategy?
 
I'm NOT asking the how. I know how to program such a strategy! In fact, my ability to do so is the motivation of this question.

I'm asking whether does third party backtesting software, for example Ninja Trader, CQG, CMC Markets, allow one to do that. Can you go up to CQG and do this "apply a moving average of every 12th data point" (which technically won't be enough because the 12th 5 min OHLC only gives you the close of the hourly not the whole hour's OHLC). I suspect not.

Since I don't know ALL the third party software out there, I don't know whether they can or can not do multi-timeframe backtesting. That's why I'm here in this forum asking that question.
 
Thanks.

Anyone else have other views? I don't want to go enthusiastically saying I can do multi-timeframe backtesting and have responses of "Oh, I do that too with XXX package."
 
Back
Top