Physical Sciences and Mathematics Commons^™

An Investigation Of Cluster Analysis, John C. Klingel

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Three cluster analysis programs were used to group the same 64 individuals, generated so as to represent eight populations of eight individuals each. Each individual had quantitative values for seven attributes. All eight populations shared a common attribute variance-covariance matrix.

The first program, from F. J. Rohlf's MINT package, implemented single linkage. Correlation was used as the basis for similarity. The results were not satisfactory, and the further use of correlation is in question.

The second program, MDISP, bases similarity on Euclidean distance. It was found to give excellent results, in that it clustered individuals into the exact populations from …

Analysis Of Case Histories By Markov Chains Using Juvenile Court Data Of State Of Utah, Soo-Hong Uh

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

The purpose of this paper is to analyze juvenile court data using Markov Chains. A computer program was generalized with a single array orientation for analyzing realizations of a Markov Chain to the k^th order within machine limitations. The data used in this paper were gathered by the Juvenile Court of the State of Utah for administrative purposes and limited to District II. The results from the paper, "Statistical Inference About Markov Chains" by Anderson and Goodman, were applied for testing hypotheses. The paper is divided into five chapters: introduction, statistical background, methodology, analysis and summary, conclusions.

The Evaluation Of Glasser's Maximum Likelihood Method On Missing Data In Regression, Gayle M. Yamasaki

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Missing data in regression is often a problem to research workers because standard regression methods are applicable only to complete data sets. At present there are three general methods for solving the problem of missing data.

At first, the reduced data method, reduces the incomplete data set to a complete data set before analyzing. Although this method is very simple to apply, substantial amounts of information are sometimes lost when data is eliminated. This results in less precise estimates of the regression parameters.

The second method, generalized least squares, estimates the missing values through least squares techniques, thus obtaining a …

Integer Programming By Cutting Planes Methods, Sung-Yen Wu

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Linear programming is a relatively new, very important branch of modern mathematics and is about twenty five years old.

In this day and age, most planners and decision makers will acknowledge that some linear optimization problems are worth the expense and trouble to solve. Using linear programming technique as a tool to make decision plannes are able to greatly redice cost or increase profit for any project under consideration.

Since Dr. George B. Dantzig published his first paper on the simplex method in 1947, progress in that field has been rapid. Although the first applications were military in nature, it …

A Nonparametric Solution For Finding The Optimum Useful Life Of Equipment, Barry T. Stoll

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

It is often the case that equipment used by industry must be replaced with new equipment from time to time either because frequent malfunctions make it too costly to repair, or because the equipment has simply worn out. The new equipment often has the nature of either malfunctioning soon after installation due to manufacturing defects, or functioning for an extended period of time because it is free of these defects. For this reason, equipment is often given a preliminary running called the burn-in which gives no useful output but merely tests for manufacturing defects. Also, after a given amount of …

Computer Programs Supporting The Teaching Of Statistics, Chien-Hwa Liu

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

During the past few years there has been an increasing interest in developing computer packages to enhance the teaching of elementary statistics. The conventional ways of teaching statistics have used such devices as desk calculators, tables of functions, short-cut calculating formulas and electronic calculators, etc. to manipulate the involved computations. Electronic computers, in the past decade, have been broadly used in universities and colleges in many ways. It is only natural to extend the use of computers to the teaching function. Remote terminals can now be installed in any classroom and bring the computer to the students.

Canonical Analysis Of Several Sets Of Variables, Chong S. Kim

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

Many procedures have been proposed for analyzing and describing multivariate dependence structure.

Partial correlation, multiple correlation, canonical correlation, principal component and factor analysis are used to analyze the dependence structures of a multinormal population.

For a partial correlation analysis it is necessary to decide which variables are to be correlated and which of the remaining variables must be held constant.

Multiple correlation demands that one variable be dependent upon some or all of the remaining variates.

For a canonical correlation, the variables must be collected into two or more sets. The factor analysis technique is for explaining the covariances of …

A Monte Carlo Comparison Of Nonparametric Reliability Estimators, Jia-Jinn Yueh

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

It is very difficult to construct a reliability model for a complex system. However, the reliability model for a series configuration is relatively simple. In the simplest case in which the components are mutually independent, the system reliability can be represented as follows:

R_s(x) = ∑ⁿ_i=1R_i(x),

where R_i is the reliability for the i^th component. It is also known that for moderate levels of system reliability for large systems, the component reliability must be high.

Extreme Value Theory indicates that under very general conditions, the initial form of the distribution function …