# The usefulness of higher order derivatives (on small time series)

**URL:** https://discourse.flucoma.org/t/the-usefulness-of-higher-order-derivatives-on-small-time-series/609
**Category:** Usage Questions
**Created:** [July 25, 2020, 12:14am UTC](https://discourse.flucoma.org/t/the-usefulness-of-higher-order-derivatives-on-small-time-series/609 "2020-07-25T00:14:58Z")
**Posts on this page:** 1
**Showing post:** 8

<div class="post-metadata">

### Author: ![rodrigo.constanzo](https://discourse.flucoma.org/user_avatar/discourse.flucoma.org/rodrigo.constanzo/32/12_2.png) [@rodrigo.constanzo](https://discourse.flucoma.org/u/rodrigo.constanzo)
#### Post date: [July 27, 2020, 9:45am UTC](https://discourse.flucoma.org/t/the-usefulness-of-higher-order-derivatives-on-small-time-series/609/8 "2020-07-27T09:45:34Z")

</div>

> [@weefuzzy](#):
>
> Second derivative is often useful, but word of caution: approximate derivatives like this are sensitive to noise in the signal and will get ‘worse’ (noise-wise) for each higher order derivative (and therefore less useful). Also, with the exeedinfly short signals you’re using, I’m dubious about how meaningful they’ll be (because we’re down to a single sample)

Indeed. That’s why I want to do [some testing](https://discourse.flucoma.org/t/regression-classification-regressification/547/40) to see if the improved results I was getting were because of higher MFCC counts or higher derivs (or both).

It’s a bit faffy to manually do that, but thankfully I have a test/script that I can run these things through and get a number back as to how accurate it was.

Depending on the results, I may revisit [your comments/thoughts about using raw melband representations](https://discourse.flucoma.org/t/onset-based-regression-jit-mfcc-example/464/29), now that I can assess things in a more scientific manner.

---

_[View the full topic](https://discourse.flucoma.org/t/the-usefulness-of-higher-order-derivatives-on-small-time-series/609)._