Home › Categories › Science & Nature› Technical Report 411

Technical Report 411.pdf

tr411.pdf

🔗 Source: stat.washington.edu

📊 Size: 207 KB

📄 Pages: 10 pages

⬇️ Downloads: 283

Summary

Researchers propose a class of Gaussian rank-r latent models for cross-covariance, showing that any variance-covariance matrix induced by rank-r reduced-rank regression can be induced by a rank-r latent model. They specify rank-r constraint models, rank-r paired latent models, and rank-r single latent models, providing a map between spaces of models. The main result, Theorem 2.1, states that for each covariance matrix in the rank-r constraint model, there is at least one set of parameter values in the rank-r symmetric single latent model that induces it. This is proven by constructing matrices U and V, which are bases for the ranges of X and Y, and using them to define matrices A and B that parameterize the covariance matrix.

Download File

Description

Researchers propose a class of Gaussian rank-r latent models for cross-covariance, showing that any variance-covariance matrix induced by rank-r reduced-rank...

Technical Information

File Format: PDF
File Size: 207 KB
Pages: 10
Language: EN
Total Downloads: 283
Last Updated: 1 week ago

Document Overview

This PDF document about Technical Report 411 provides comprehensive information and guidance. Whether you're a beginner or advanced user, this resource offers valuable insights into Technical Report 411.