A characterization of the matrix variate normal distribution having identically distributed row vectors based on conditional normality is given. 1997 Academic Press ## 1. INTRODUCTION AND BASIC RESULTS Let X 1 and X 2 be two identically distributed random variables. Suppose that X 1 | X 2 =x 2 has a
Characterization of matrix variate normal distributions
โ Scribed by A.K Gupta; T Varga
- Publisher
- Elsevier Science
- Year
- 1992
- Tongue
- English
- Weight
- 311 KB
- Volume
- 41
- Category
- Article
- ISSN
- 0047-259X
No coin nor oath required. For personal study only.
๐ SIMILAR VOLUMES
It is well known that i.i.d. (independent and identically distributed) normal random variables are transformed into i.i.d. normal random variables by any orthogonal transformation. Less well known are nonlinear transformations with the above-mentioned property. In this work we present nonlinear tran
It is a well-known result (which can be traced back to Gauss) that the only translation family of probability densities on \(\mathbb{R}\) for which the arithmetic mean is a maximum likelihood estimate of the translation parameter originates from the normal density. We generalize this characterizatio