The Cox regression model, random censoring and locally optimal rank tests (Q1061427)
From MaRDI portal
scientific article
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | The Cox regression model, random censoring and locally optimal rank tests |
scientific article |
Statements
The Cox regression model, random censoring and locally optimal rank tests (English)
0 references
1984
0 references
The paper considers the Cox regression model for survival data. The parameters of interest are contained in the vector \(\beta\) and the hypothesis of interest is \(H_ 0:\beta =0.\) It is shown that the log- rank statistic does not generally provide LMPR (or maximin) tests, even for the case of the two sample problem. For this reason conditions on the hazard rates are investigated, under which the log-rank statistic has some locally optimal properties. The paper addresses also to the gain in efficiency of the Cox procedure over the log-rank procedure, and some asymptotic results are given.
0 references
random censoring
0 references
locally optimal rank tests
0 references
covariates
0 references
LMPR test
0 references
withdrawal
0 references
locally most powerful rank test
0 references
Cox regression model
0 references
survival data
0 references
log-rank statistic
0 references
hazard rates
0 references
