A LINEAR COMPOSITE METHOD FOR TEST SCORES WITH MISSING VALUES

1999 Fiscal Year Final Research Report Summary

• Project/Area Number: 10680315
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Statistical science
• Research Institution: NIIGATA UNIVERSITY
• Principal Investigator: SHIBAYAMA Tadashi NIIGATA UNIVERSITY DEPARTMENT OF EDUCATION AND HUMAN SCIENCES, 教育人間科学部, 助教授 (70240752)
• Project Period (FY): 1998 – 1999
• Keywords: INCOMPLETE DATA / MISSING VALUES / COMPOSITE SCORES / INDIVIDUAL DIFFERENCES / MANOVA / ENTRANCE EXAMINATION / CANANICAL ANALYSIS / 大学入試

Research Abstract

Here I present a linear composite method for incomplete data. The method is based on the least squares criterion and can obtain solutions analytically in such a way that is individual differences are maximized allowing for differences between means of variables. If a data matrix is complete, the method is equivalent to ordinary PCA. A simulation study is given in order to demonstrate the efficacy of the method.