Statistical inference on comparing two distribution functions with a possible crossing point
Based on rank statistics, new distribution-free tests on the crossing point between two distribution functions are proposed. Simulation indicates that the new tests have higher power than tests in the literature. A new point estimate of the crossing point is given and is shown to be strongly consistent and asymptotically normal. Two kinds of asymptotically distribution-free confidence intervals of the crossing point have also been given.
Year of publication: |
2002
|
---|---|
Authors: | Chen, Guijing ; Chen, Jiahua ; Chen, Yuming |
Published in: |
Statistics & Probability Letters. - Elsevier, ISSN 0167-7152. - Vol. 60.2002, 3, p. 329-341
|
Publisher: |
Elsevier |
Keywords: | Rank statistics Distribution-free test Distribution-free confidence interval Asymptotic distribution |
Saved in:
Saved in favorites
Similar items by person
-
A new Lorenz-type hyperchaotic system with a curve of equilibria
Chen, Yuming, (2015)
-
On extending the Brunk-Prokhorov strong law of large numbers for martingale differences
Hu, Shuhe, (2008)
-
Small Area Quantile Estimation
Chen, Jiahua, (2018)
- More ...