Exploratory factor analysis-Parameter estimation and scores prediction with high-dimensional data

JOURNAL OF MULTIVARIATE ANALYSIS

卷:148

Article

Sundberg, Rolf¹  Feldmann, Uwe² 

[1] Stockholm Univ, Math Stat, SE-10691 Stockholm, Sweden

[2] Univ Saarland, Med Biometry Epidemiol & Med Informat, D-66123 Saarbrucken, Germany

关键词: EFA;    FA;    Factor score estimation;    Fixed point iteration;    Likelihood equations;    More variables than observations;    SVD;   

DOI  :  10.1016/j.jmva.2016.02.013

来源: Elsevier

【 摘 要 】

In an approach aiming at high-dimensional situations, we first introduce a distribution-free approach to parameter estimation in the standard random factor model, that is shown to lead to the same estimating equations as maximum likelihood estimation under normality. The derivation is considerably simpler, and works equally well in the case of more variables than observations (p > n). We next concentrate on the latter case and show results of type: Albeit factor, loadings and specific variances cannot be precisely estimated unless n is large, this is not needed for the factor scores to be precise, but only that p is large; A classical fixed point iteration method can be expected to converge safely and rapidly, provided p is large. A microarray data set, with p = 2000 and n = 22, is used to illustrate this theoretical result. (c) 2016 The Authors. Published by Elsevier Inc.

【 授权许可】

Free