Probability matrices, non-negative rank, and parameterization of mixture models (Q975609): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Set profile property. |
||
| Property / MaRDI profile type | |||
| Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Revision as of 02:48, 5 March 2024
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Probability matrices, non-negative rank, and parameterization of mixture models |
scientific article |
Statements
Probability matrices, non-negative rank, and parameterization of mixture models (English)
0 references
10 June 2010
0 references
The authors parameterize matrices of sum one and rank at most two using the least possible number of parameters. They also show how this parameterization relates to a class of statistical models, known in probability and statistics as mixture models for contingency tables. In particular, they show how to use this parameterization to make some optimization problems computationally easier.
0 references
contingency tables
0 references
optimization on algebraic varieties
0 references
parameter redundancy
0 references
mixture models
0 references
