Accession Number : AD0853245

Title :   A Decision Theoretic Approach to the Stochastic Approximation of the Root of an Unknown Function.

Descriptive Note : Master's thesis,

Corporate Author : AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING

Personal Author(s) : Emmerichs, Robert M.

Report Date : JUN 1969

Pagination or Media Count : 54

Abstract : A decision theoretic approach is used to characterize stochastic approximation procedures. The basic foundation of stochastic approximation and statistical decision theory is presented. These two fields are combined using the restrictive problem of estimating the unknown root of a linear function with unit slope. The restrictions on the function are relaxed, and convergence almost surely is demonstrated in the general case. A section on applications of stochastic approximation in fields related to systems analysis is presented. (Author)

Descriptors :   *DECISION THEORY), (*STOCHASTIC PROCESSES, ITERATIONS), (*ITERATIONS, PROBLEM SOLVING, CONVERGENCE, APPROXIMATION(MATHEMATICS), REGRESSION ANALYSIS, CONTROL SYSTEMS, THESES.

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE