Award Date

1-1-2003

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Mathematical Sciences

First Committee Member

Hokwon Cho

Number of Pages

48

Abstract

For most non-parametric statistical inference, the sampling theory of order statistics has been playing a fundamental role because of that properties of the range and the average of the smallest and largest order statistics are useful for estimate the population parameters of both large and small samples; The applications of the method using order statistics in a given sample to estimate the parametric values appear quite often in the literature. For a certain data set, such as stock market data, the method we are considering may have an advantage in estimating the mean due to the fact that the stock data have a fairly large amount of observations during a given period, even a day; In addition to estimating the mean, it is of interest to compute (1-alpha) 100% confidence limits as well. Using the two extremes, X(1) and X(n), we wish to construct a confidence interval for the population mean mu.

Keywords

Estimation; Extremes; Mean; Order; Population; Statistics; Two

Controlled Subject

Mathematics; Statistics

File Format

pdf

File Size

819.2 KB

Degree Grantor

University of Nevada, Las Vegas

Language

English

Permissions

If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to digitalscholarship@unlv.edu and include clear identification of the work, preferably with URL.

Identifier

https://doi.org/10.25669/uf1o-6u27


Share

COinS