Asymptotic Statistics pp 405-413 | Cite as
Conditional Rank Tests for the Two-Sample Problem with Partially Observed Data
Conference paper
Abstract
In the two-sample random censorship model with the additional complication that the minimum variable is observable only for the uncensored data we develop asymptotically optimal conditional rank tests for testing the null hypothesis of randomness H 0 being finite sample distribution free under H o
AMS 1991 subject classifications
62G20 62G10Key words
Censoring partially observed data conditioning finite sample validity asymptotic normality optimality.Preview
Unable to display preview. Download preview PDF.
References
- Albers, W. and Akritas, M.G. (1987). Combined rank tests for the two-sample problem with randomly censored data. J. American Statist. Ass., 82, 648–655.MathSciNetMATHCrossRefGoogle Scholar
- Hájek, J. and Sidák, Z. (1967). Theory of Rank Tests. Academic Press, New York.MATHGoogle Scholar
- Janssen, A. (1989). Local asymptotic normality for randomly censored models with applications to rank tests. Statistica Neerlandica 43, 109–125.MathSciNetMATHCrossRefGoogle Scholar
- Neuhaus, G. (1988). Asymptotically optimal rank tests for the two-sample problem with randomly censored data. Communications in Statistics-Theory and Methods 17, 2037–2058.MATHCrossRefGoogle Scholar
- Suzuki, K. (1985). Nonparametric estimation of lifetime distributions from a record of failures and follow-ups. J. American Statist. Ass. 80, 68–72.MATHCrossRefGoogle Scholar
- Witting, H. (1985). Mathematische Statistik I. Teubner, Stuttgart.MATHGoogle Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 1994