In a clinical trial to compare two treatments, suppose that patients are assigned to treatment according to a certain class of restricted randomization rules. Conditional on the difference in numbers between the two treatment groups, the asymptotic null distribution of a class of permutation tests is shown to be Gaussian.
Publié le : 1988-09-14
Classification:
Conditional inference,
local limit theorem,
randomization model,
restricted randomization rules,
60F05,
62G10
@article{1176350952,
author = {Smythe, R. T.},
title = {Conditional Inference for Restricted Randomization Designs},
journal = {Ann. Statist.},
volume = {16},
number = {1},
year = {1988},
pages = { 1155-1161},
language = {en},
url = {http://dml.mathdoc.fr/item/1176350952}
}
Smythe, R. T. Conditional Inference for Restricted Randomization Designs. Ann. Statist., Tome 16 (1988) no. 1, pp. 1155-1161. http://gdmltest.u-ga.fr/item/1176350952/