Matrix-valued distributions are used in continuous multivariate analysis to model sample datamatrices of continuous measurements; their use seems to be neglected for binary, or more generallycategorical, data. In this paper we propose a matrix-valued Bernoullidistribution, based on the log-linear representation introducedby Cox (1972) for the Multivariate Bernoulli distribution with correlated components.
|Number of pages||13|
|Journal||Journal of Multivariate Analysis|
|Publication status||Published - 2006|
All Science Journal Classification (ASJC) codes
- Statistics and Probability
- Numerical Analysis
- Statistics, Probability and Uncertainty