Adaptive Robust Estimation of Location and Scale Parameters of Symmetric Populations.
Abstract
Critical values of several criteria, for equal Type I and Type II error rates (probabilities of misclassification), have now been established for sample sizes n = 8 (4) 24 by means of a Monte Carlo study based on 5000 random samples of each size from each of the above populations Uniform (U), Normal (N), and Double (D). Mean square errors of the adaptive estimates are compared with those of the ML estimates if the population from which each sample came is known, and the effect of debiasing the ML estimates of the standard deviation (sigma) is studied. Adaptive estimation of the canonical scale parameter F sigma, where the factor F is defined as the multiplier of sigma such that F sigma is the 97.5% point of a population symmetric about zero, is also considered. Monte Carlo studies have also been conducted to determine the performance of the various criteria when applied to an independent set of random samples (obtained by using a different seed for the random number generator) from U, N and D and to random samples from several other symmetric populations, for the above values of n and for the intermediate values n = 10 (4) 22, with critical values of the criteria determined by interpolation.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1978
- Accession Number
- ADA062436
Entities
People
- Albert H. Moore
- H. Leon Harter
- Thomas F. Curry
Organizations
- Flight Dynamics Laboratory