The Law of the Iterated Logarithm and Central Limit Theorem for L-Statistics.
Abstract
The main idea in this paper is that we devise an effective way of combining the Smirnov's law of the iterated logarithm for empirical processes, and some well-known results of limit behavior of L-statistics to establish new results on the central limit theorem, law of the iterated logarithm, and strong law of large numbers, for L-statistics. We show further that this approach can be pursued profitably to obtain necessary and sufficient conditions for either almost sure convergence or convergence in distribution of some well-known L-statistics and U-statistics. A law of the logarithm for weighted sums of order statistics is stated with no proof.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1997
- Accession Number
- ADA328387
Entities
People
- Deli Li
- M. B. Rao
- R. J. Tomkins
Organizations
- Pennsylvania State University