Accession Number : AD0657463

Title :   INADMISSIBILITY OF VARIOUS 'GOOD' STATISTICAL PROCEDURES WHICH ARE TRANSLATION INVARIANT.

Descriptive Note : Technical rept.,

Corporate Author : MICHIGAN STATE UNIV EAST LANSING DEPT OF STATISTICS

Personal Author(s) : Perng,Shian-Koong

Report Date : 31 AUG 1967

Pagination or Media Count : 44

Abstract : The purpose of this thesis is to try to show that certain moment conditions are essential for the admissibility of various 'good' translation invariant statistical procedures. Section 3 indicates the inadmissibility of a unique best translation invariant estimate if a certain moment condition fails to be satisfied. Section 4 and 5 shows, by example, the inadmissibility of a best translation invariant test if either the test is non-unique or certain moment conditions fail to hold. Section 6 shows, by example, that a certain translation invariant confidence interval is inadmissible if a certain moment condition fails to hold.

Descriptors :   (*STATISTICAL PROCESSES, INVARIANCE), SAMPLING, DISTRIBUTION FUNCTIONS, TRANSFORMATIONS(MATHEMATICS), THESES, THEOREMS, INEQUALITIES

Subject Categories : Statistics and Probability

Distribution Statement : APPROVED FOR PUBLIC RELEASE