Hi!

I am developing a trading strategy based on trading Nasdaq 100 Index.

Previously, I was using QQQ for strategy building and optimization as QQQ is tracking the Nasdaq 100 Index.

However, suddenly, I am thinking if I could use ^NDX directly as ^NDX is the Nasdaq 100 Index itself and has a much longer history.

QQQ has a history back to the year of 1999 while ^NDX has a history back to the year of 1985.

Ideally, using ^NDX for my strategy building and optimization will be better as it has much longer history than QQQ.

But, my question is if I would use ^NDX for my strategy building and optimize, but in fact (in the real world), I will use QQQ to make real trade, will this approach work?

QQQ is supposed to track ^NDX. But is it tracking ^NDX at 100%? If it does track ^NDX at 100%, I should be able to use the above approach (use ^NDX for my strategy building and optimization then use QQQ for real trade), right?

Does Wealth Lab Dev provide any tool to compare ^NDX and QQQ to see if they are correlated at 100% or how much they are correlated?

Could somebody shed some light on this?

Thanks!


Alex

Hi Alex,

Of course WLD provides correlation. There are two members of Community Indicators which return Pearson's correlation, CorrelationXL and Correlation:

Correlation

They're doing the same thing - just were coded by different people.

Thank you, Eugene!

I have used the following code sample to demo the correlation calculation:

CODE:
Please log in to see this code.

It works and it calculates correlation on CMO and RSI value of any one instrument.

However, my plan is to read two instruments (^NDX and QQQ) into a script and calculate the correlation.

Could you shed some light on how to read two instruments into a script for calculation?



Alex

Sure thing. Here goes:

CODE:
Please log in to see this code.

Pick whichever series you want (CorrelationXL or Correlation), they're spitting out the same number.

Excellent! Eugene...

The code that you provided works and from the output result, I am very sure that ^NDX and QQQ are very much correlated.

Thanks again!


Alex

Any time. Alex, you might also want to adjust the Period via exposing a parameter slider to review the correlation over different time spans.

Yes and good suggestion!

I did adjust the Period from 252 (one year) to various numbers (multiple years) to look back plus span different timeframe then calculate the correlation between ^NDX and QQQ...