Missing values in principal component analysis

被引:95
作者
Grung, B [1 ]
Manne, R [1 ]
机构
[1] Univ Bergen, Dept Chem, N-5007 Bergen, Norway
关键词
principal component analysis; missing elements; regression approach; imputation;
D O I
10.1016/S0169-7439(98)00031-8
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Calculation schemes for principal component analysis are considered for the case when some matrix elements are missing. Iterative solutions are proposed-either a set of multilinear regression problems or as singular-value decomposition problems with iterative imputation of missing values. If mean values are subtracted from the data matrix, they should also be included in the iteration scheme. Test calculations using Matlab show that the regression approach is somewhat faster than the imputation approach. The results with a substantial amount of missing data are different and superior to those obtained with the naive NIPALS algorithm in common use in chemometrics. (C) 1998 Elsevier Science B.V. All rights reserved.
引用
收藏
页码:125 / 139
页数:15
相关论文
共 15 条
  • [1] Christoffersson A, 1970, THESIS UPPSALA U
  • [2] AN APPLICATION OF FACTOR-ANALYSIS WITH MISSING DATA
    DELIGNY, CL
    NIEUWDORP, GHE
    BREDERODE, WK
    HAMMERS, WE
    VANHOUWELINGEN, JC
    [J]. TECHNOMETRICS, 1981, 23 (01) : 91 - 95
  • [3] LOWER RANK APPROXIMATION OF MATRICES BY LEAST-SQUARES WITH ANY CHOICE OF WEIGHTS
    GABRIEL, KR
    ZAMIR, S
    [J]. TECHNOMETRICS, 1979, 21 (04) : 489 - 498
  • [4] HARIGAN JA, 1975, CLUSTER ALGORITHMS, P351
  • [5] HARSHMAN RA, 1994, COMP STAT DATA ANAL, V18
  • [6] The wave mechanics of an atom with a non-Coulomb central field Part I theory and methods
    Hartree, DR
    [J]. PROCEEDINGS OF THE CAMBRIDGE PHILOSOPHICAL SOCIETY, 1928, 24 : 89 - 110
  • [7] Weighted least squares fitting using ordinary least squares algorithms
    Kiers, HAL
    [J]. PSYCHOMETRIKA, 1997, 62 (02) : 251 - 266
  • [8] KRIJNEN WP, 1995, COMPUTATION STAT, V10, P299
  • [9] Little RJ., 1995, Handbook of statistical modeling for the social and behavioral sciences, P39
  • [10] LITTLE RJA, 1987, STAT ANLA MISSING DA