Inequalities for Joint Distributions of Quadratic Forms.
VIRGINIA POLYTECHNIC INST AND STATE UNIV BLACKSBURG
Pagination or Media Count:
Chebychev inequalities are given for joint central and noncentral distributions of k quadratic forms these are sharpened when k 2 using the canonical correlations of Hotelling. Complementary inequalities are found as versions of Markovs inequality. Applications are noted in ballistics, in statistical quality control, in establishing consistency of Gauss-Markov estimates under dependence, and in constructing conservative joint confidence sets depending on the underlying distribution only through its low order moments. Author
- Statistics and Probability