On a separation theorem for generalized eigenvalues and a problem in the analysis of sample surveys (Q1072277)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | On a separation theorem for generalized eigenvalues and a problem in the analysis of sample surveys |
scientific article |
Statements
On a separation theorem for generalized eigenvalues and a problem in the analysis of sample surveys (English)
0 references
1985
0 references
Methods have been developed earlier for analyzing multiway tables of categorical data in complex sample surveys which use log-linear models. In this paper bounds are obtained for the asymptotic percentage points of chi-squared tests of fit for these log-linear models. These bounds are obtained by using certain separation inequalities for the generalized eigenvalues of a matrix X'AX with respect to a matrix X'BX, when both A and B are assumed to be non-negative definite.
0 references
Poincaré separation theorem
0 references
multiway tables of categorical data
0 references
complex sample surveys
0 references
log-linear models
0 references
bounds
0 references
asymptotic percentage points
0 references
chi-squared tests of fit
0 references
separation inequalities
0 references
generalized eigenvalues
0 references
0 references
0 references
0 references
0 references
0 references
