Screening procedures using quadratic forms (Q3703153)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3930190
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Screening procedures using quadratic forms |
scientific article; zbMATH DE number 3930190 |
Statements
Screening procedures using quadratic forms (English)
0 references
1985
0 references
bivariate chi-square
0 references
bivariate F
0 references
product acceptance
0 references
one-dimensional performance variable
0 references
lower specification limit
0 references
upper specification limit
0 references
quadratic form
0 references
normal distribution
0 references
screening procedure
0 references
linear combination of the screening variables
0 references
0.8382079005241394
0 references
0.810339093208313
0 references
0.8094643354415894
0 references
0.8051598072052002
0 references
An item is acceptable if its one-dimensional performance variable Y is greater than a lower specification limit and less than an upper specification limit. When measuring of Y is destructive or too expensive it may be reasonable to use a p-dimensional screening variable X which is positively correlated to Y. An item is selected if a quadratic form in X is sufficiently small. For this it is assumed that X, Y follow a \((p+1)\)-dimensional normal distribution with known parameters. Then the screening procedure is determined such that the proportion of acceptable items after screening is maximized for a fixed screening proportion. Some of the possible cases of unknown parameters are considered. Finally some hints are given for replacing the quadratic form in X by a linear combination of the screening variables.
0 references