# Strange correlation behavior
**URL:** https://forum.numer.ai/t/strange-correlation-behavior/4837
**Category:** Data Science
**Created:** [January 22, 2022, 5:44pm UTC](https://forum.numer.ai/t/strange-correlation-behavior/4837 "2022-01-22T17:44:20Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@mrquantsalot](https://forum.numer.ai/u/mrquantsalot)
#### Post date: [January 22, 2022, 5:44pm UTC](https://forum.numer.ai/t/strange-correlation-behavior/4837/1 "2022-01-22T17:44:20Z")
I was watching this NNTaleb video on correlation ([https://www.youtube.com/watch?v=o9Ac85xdjE4](https://www.youtube.com/watch?v=o9Ac85xdjE4)) and he talks about how correlation is often not a good metric for measuring dependence between variables.
Here’s the example:
Any nonlinear model can use x to predict y. The takeaway could be not to use correlation to decide what features to include.
---
### Author:  [@of\_s](https://forum.numer.ai/u/of_s)
#### Post date: [January 22, 2022, 8:08pm UTC](https://forum.numer.ai/t/strange-correlation-behavior/4837/2 "2022-01-22T20:08:39Z")
It is critical to discern between correlation and dependence…
[https://cran.r-project.org/web/packages/NNS/vignettes/NNSvignette\_Correlation\_and\_Dependence.html](https://cran.r-project.org/web/packages/NNS/vignettes/NNSvignette_Correlation_and_Dependence.html)
---
### Author:  [@gammarat](https://forum.numer.ai/u/gammarat)
#### Post date: [January 22, 2022, 8:45pm UTC](https://forum.numer.ai/t/strange-correlation-behavior/4837/3 "2022-01-22T20:45:19Z")
> [@mrquantsalot](#):
>
> The takeaway could be not to use correlation to decide what features to include.
I think a better takeaway is just that correlation is limited and should be used judiciously. If you take your triangle example (or Taleb’s—thanks for posting the video, btw) you’ll note that while a single correlation doesn’t produce any useful information, two correlations (one on each leg of the triangle) would. That then introduces a new question, how to partition the domain under analysis into suitable “regimes” where simple methods suffice.
The regime question surfaces here from time to time, and it’s one I do find fascinating. In practical terms, one might think of regimes in the Tournament as eras in which a specific set of features might correlate well with the targets, while a different regime would consist of eras in which a different set of features would do so. If one could identify the regime of an era before inverting the features to estimating targets, then Bob’s-yer-uncle you’ll be rich 😆.