ESTIMATION OF THE SCALE PARAMETER OF THE GAMMA DISTRIBUTION BY USE OF L ORDER STATISTICS.
AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OHIO SCHOOL OF ENGINEERING
Pagination or Media Count:
A technique is developed for estimating the scale parameter of a gamma distribution with known shape parameter using L order statistics. A best linear unbiased estimate may be computed by applying tabulated multipliers to L of the first M ordered observations. The variance of an estimator using all M of the observations and the efficiencies of the L-order-statistic -estimators are given. Tabled ranges include shape parameters of alpha 116 for sample sizes of N 1115. A method of obtaining the multipliers when the shape parameter is not an integer is also shown. Estimation by the use of L-order-statistics is particularly useful, for highly efficient estimators relative to the M-order-statistic-estimators may be obtained with a significant reduction in computational effort. Author
- Statistics and Probability
- Manufacturing and Industrial Engineering and Control of Production Systems