Bound on the Classification Error for Discriminating Between Multivariate Populations with Specified Means and Covariance Matrices
Isii, K. ; Taga, Y.
Ann. Statist., Tome 6 (1978) no. 1, p. 132-141 / Harvested from Project Euclid
Let $\mathscr{F}_1, \mathscr{F}_2$ be two families of $p$-variate distribution functions with specified means $\mathbf{\mu}_i (i = 1,2)$ and nonsingular covariance matrices $\Sigma_i$, and let $\pi_i$ be the prior probability assigned to $\mathscr{F}_i$ for $i = 1, 2$. The objective is to discriminate whether an observation $\mathbf{x}$ is from a distribution $F_1 \in \mathscr{F}_1$ or $F_2 \in \mathscr{F}_2$. Given a pair $F = (F_1, F_2)$ the error probability for classification rule $\phi$ is denoted by $e(\phi, F)$. In this paper the values of $\sup_F \inf_\phi e(\phi, F)$ and $\inf_\phi \sup_F e(\phi, F)$ are found and conditions for the existence of a saddle point of $e(\phi, F)$ are given. Also a saddle point is found when it exists. When $\phi$ is restricted to linear classification rules the same problems are considered. The mathematical programming method for finding a saddle point is also outlined.
Publié le : 1978-01-14
Classification:  Discrimination,  classification rule,  bound for error probability,  minimax theorem,  62H30,  62G99,  90C05
@article{1176344072,
     author = {Isii, K. and Taga, Y.},
     title = {Bound on the Classification Error for Discriminating Between Multivariate Populations with Specified Means and Covariance Matrices},
     journal = {Ann. Statist.},
     volume = {6},
     number = {1},
     year = {1978},
     pages = { 132-141},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176344072}
}
Isii, K.; Taga, Y. Bound on the Classification Error for Discriminating Between Multivariate Populations with Specified Means and Covariance Matrices. Ann. Statist., Tome 6 (1978) no. 1, pp.  132-141. http://gdmltest.u-ga.fr/item/1176344072/