# Time Series with the data

**URL:** <https://datachallenge.cfm.fr/t/time-series-with-the-data/75>\
**Category:** Modeling\
**Created:** [April 7, 2018, 11:10am UTC](https://datachallenge.cfm.fr/t/time-series-with-the-data/75 "2018-04-07T11:10:08Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mayank\_More](https://avatars.discourse-cdn.com/v4/letter/m/f475e1/32.png) [@Mayank\_More](https://datachallenge.cfm.fr/u/Mayank_More)\
**Post date:** [April 7, 2018, 11:10am UTC](https://datachallenge.cfm.fr/t/time-series-with-the-data/75/1 "2018-04-07T11:10:08Z")

</div>

Since the days are randomized, we can not use time series model such as ARCH/GARCH to predict the volatility. Is anyone using such models after doing some transformation on the data or only regression models and its variation can be tried out at this data?

---

<div class="post-metadata">

**Author:** ![lebigot](https://yyz1.discourse-cdn.com/flex029/user_avatar/datachallenge.cfm.fr/lebigot/32/10_2.png) [@lebigot](https://datachallenge.cfm.fr/u/lebigot)\
**Post date:** [April 16, 2018, 6:48pm UTC](https://datachallenge.cfm.fr/t/time-series-with-the-data/75/2 "2018-04-16T18:48:37Z")

</div>

It is in principle possible to use time series models, at least within a single day. Now, how much this provides good performance remains to be seen.

Hopefully those who have done this can share their experience and get feedback in return! There are some interesting ideas that can be tried based on models.