Doctoral Dissertations

Abstract

"A "Schmidt filter" is proposed to compute an optimal orthonormal basis for a set of "noisy" filter input functions. Procedures for determining the transfer function and inverse transfer function of the filter are given. The Schmidt filter is applied to the problem of determining mathematical models of discrete, stationary, linear, dynamic systems for the case where measurements may be corrupted by noise of unknown statistics. The identification problem is reconsidered for the case where noise and signal moments are specified. Procedures are given which insure unbiased, adaptive estimates of system order and parameters for this case. These theoretical propositions are applied to the modeling of speculative prices. The stock market is formulated as a discrete, linear, dynamic system and the results of several simulation studies are presented. Evidence indicates that certain segments of the market can be approximated by high-order linear systems computed from small samples and tends to refute the random walk hypothesis. Computer programs (written in PL/1) are presented which allow for efficient digital realization of the theoretical procedures discussed in the body of this work"--Abstract, page iii.

Advisor(s)

Flanigan, V. J.

Committee Member(s)

Johnson, R. T. (Richard T.)
Pazdera, John S., 1941-1974
Ho, C. Y. (Chung You), 1933-1988
Sieck, Lawrence K.
Faucett, T. R.

Department(s)

Mechanical and Aerospace Engineering

Degree Name

Ph. D. in Mechanical Engineering

Publisher

University of Missouri--Rolla

Publication Date

1972

Pagination

xi, 171 pages

Note about bibliography

Includes bibliographical references (pages 139-145).

Rights

© 1972 Allen Glenn Behring, All rights reserved.

Document Type

Dissertation - Open Access

File Type

text

Language

English

Library of Congress Subject Headings

Securities -- Prices -- Mathematical models
Stock exchanges -- Computer simulation
Speculation -- Computer simulation

Thesis Number

T 2755

Print OCLC #

6034214

Electronic OCLC #

887720979

Share

 
COinS