Activity Number:
|
465
- SPEED: Statistical Computing: Methods, Implementation, and Application, Part 1
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, July 31, 2019 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistical Computing
|
Abstract #307197
|
|
Title:
|
Tensor Variate Models Applied to Sensor Data
|
Author(s):
|
Peter Tait* and Paul D McNicholas
|
Companies:
|
McMaster University and McMaster University
|
Keywords:
|
child health;
physical activity;
accelerometers;
tensors
|
Abstract:
|
Motivated by our work in pediatric research, a tensor variate model is proposed to cluster tensor variate data. The model builds on the Tucker tensor decomposition and which allows the parameters to be easily interpretable. The model parameters are estimated using an EM algorithm. Our model is implemented in Julia, a dynamic and high performance numerical computing language. The tensor variate data is generated from accelerometers worn by a cohort of school aged children.
|
Authors who are presenting talks have a * after their name.