# Cumulation of rank data (like quantiles and mean)

**URL:** <https://datashield.discourse.group/t/cumulation-of-rank-data-like-quantiles-and-mean/399>\
**Category:** Developer support\
**Created:** [1 June 2021 09:33 UTC](https://datashield.discourse.group/t/cumulation-of-rank-data-like-quantiles-and-mean/399 "2021-06-01T09:33:17Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![rapus95](https://avatars.discourse-cdn.com/v4/letter/r/eb9ed0/32.png) [@rapus95](https://datashield.discourse.group/u/rapus95)\
**Post date:** [1 June 2021 09:33 UTC](https://datashield.discourse.group/t/cumulation-of-rank-data-like-quantiles-and-mean/399/1 "2021-06-01T09:33:18Z")

</div>

I just stumbled over a method/structure for combining multiple sources of ranked data: [GitHub - tdunning/t-digest: A new data structure for accurate on-line accumulation of rank-based statistics such as quantiles and trimmed means](https://github.com/tdunning/t-digest) I wonder if these might be of greater interest for datashield, given that for example there’s currently no summary method that combines the data from all sources.

---

<div class="post-metadata">

**Author:** ![PatRyserWelch](https://yyz2.discourse-cdn.com/free1/user_avatar/datashield.discourse.group/patryserwelch/32/11_2.png) [@PatRyserWelch](https://datashield.discourse.group/u/PatRyserWelch)\
**Post date:** [15 June 2021 08:58 UTC](https://datashield.discourse.group/t/cumulation-of-rank-data-like-quantiles-and-mean/399/2 "2021-06-15T08:58:52Z")

</div>

Hi Thank you. I am currently working on such summary technique. It is useful.

P.