Webinar: Multy-way CoDa (2022-10-26)

WEBINAR

Multi-way compositional data: coordinates, interpretation, and open challenges

Violetta Simonacci

To get access to the video recording of the talk, please, login and go to menu Members area -> Video Recording & slides

  • Day: 26-10-2022
  • Time: 12:00h (CEST: UTC/GMT +2 h)
  • Link: https://us06web.zoom.us/j/88591669413?pwd=REZDc0tmK2VhNlkrUGJwY1FVNm93Zz09
  • Speaker: Violetta Simonacci, PhD,Assistant Professor in Statistics, University of Naples Federico II - Dept. of Social Sciences.
  • Moderator: Javier Palarea-Albaladejo
  • Title: Multi-way compositional data: coordinates, interpretation, and open challenges
  • Abstract:  In multidimensional tensor analysis, researchers are still looking for ways to refine the process of parameter estimation to approximate the interaction between and within dimensions.  In detail, the goal is to improve computational efficiency, ease of interpretability, and accuracy of results while dealing with estimation challenges connected to large datasets, pre-processing limitations, bad initialization, incorrect model specification, collinearity, and high noise. In this setting, the presence of a CoDa structure or, in general, the detection of relative variability patterns across dimensions,  exacerbate these known difficulties by adding a discussion on the choice of basis and its correct interpretation. This seminar provides an overview of the recent work conducted on higher-order CoDa tensors by focusing on the problem of coordinate identification, interpretational issues, and computational challenges.
  • Keywords:  Compositional data, tensor analysis, logratio, multi-way data

 

---------------------------------------------------------------------------------------------------

Please note that the webinar will be recorded. Feel free to forward this message to your colleagues who may be interested in attending.

Check your local time at https://www.timeanddate.com/worldclock/converter.html?iso=20221026T100000&p1=31&p2=179&p3=152&p4=101