Partial modal tensor regression proposed for heavy-tailed responses
Original title: Partial Modal Tensor Regression
The method uses a partial modal tensor covariance to reduce dimension before fitting regression on the conditional mode.
The paper proposes partial modal tensor regression for responses that are asymmetric, heavy-tailed or contaminated by outliers. A partial modal tensor covariance guides dimension reduction, and a modal regression is fit in the reduced space. Authors state consistency and asymptotic properties under regularity conditions and use a data-driven rule to pick dimension along each mode. No deployment or security results are reported.
- Existing tensor regression methods mainly target conditional means or quantiles.
- A data-driven procedure selects the effective dimension along each tensor mode.
From the original article
1 min read at arxiv.orgWith the rapid development of modern scientific research, tensor data have become increasingly prevalent across a wide range of fields. How to effectively characterize the relationship between tensor predictors and a response while preserving the intrinsic structure of tensor data has therefore become an important problem in high-dimensional statistical analysis.
By Jintao Han · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.