Large sample inference for diagnostic normal limits in gaussian populations (Q3713340)
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 3942737
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Large sample inference for diagnostic normal limits in gaussian populations |
scientific article; zbMATH DE number 3942737 |
Statements
Large sample inference for diagnostic normal limits in gaussian populations (English)
0 references
1985
0 references
tolerance interval
0 references
chi distribution
0 references
diagnosis
0 references
mean-square-error consistent estimators
0 references
large sample test
0 references
equality of Gaussian population normal limits
0 references
rate of convergence
0 references
nerve conduction studies
0 references
percentiles
0 references
Welch's test
0 references
Behrens-Fisher problem
0 references
pooled estimator
0 references
0.728938639163971
0 references
0.7126255035400391
0 references
0.7126253247261047
0 references
0.7121573686599731
0 references
For a \(N(\mu,\sigma^ 2)\) population, several estimators of the percentiles \(\mu +k\sigma\), where k is a known constant, are compared in terms of their finite-sample and asymptotic properties. A test which generalizes Welch's test for the Behrens-Fisher problem is proposed to test \(H:\mu_ 1+k\sigma_ 1=\mu_ 2+k\sigma_ 2,\) when the ith population is \(N(\mu_ i,\sigma^ 2_ i),\) \(i=1,2\). This test is compared with the standard test one would use to test H under the assumption \(\sigma_ 1=\sigma_ 2.\) A pooled estimator of the common percentile of the two populations, to be used when H is accepted, is compared with estimators one would use when \(\mu_ 1=\mu_ 2\) and \(\sigma_ 1=\sigma_ 2\). Special attention is paid to the case of \(k=2\) in view of applications in clinical medicine.
0 references