Premium
ESTIMATION IN FACTOR ANALYSIS UNDER VARIOUS INITIAL ASSUMPTIONS
Author(s) -
Lawley D. N.
Publication year - 1958
Publication title -
british journal of statistical psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.157
H-Index - 51
eISSN - 2044-8317
pISSN - 0950-561X
DOI - 10.1111/j.2044-8317.1958.tb00186.x
Subject(s) - simple (philosophy) , a priori and a posteriori , oblique case , estimation , factor (programming language) , maximum likelihood , object (grammar) , computer science , arc (geometry) , iterative method , mathematics , mathematical optimization , algorithm , econometrics , statistics , engineering , geometry , artificial intelligence , philosophy , linguistics , systems engineering , epistemology , programming language
The object of the following paper is to investigate how far it is practicable to adopt the principle of maximum likelihood for the estimation of factor loadings, when some of the loadings are assumed a priori to be zero. Both the orthogonal and the oblique case arc considered; and the solutions obtained are illustrated by applying them to a simple example. Owing to the need for an iterative procedure the labour involved increases enormously as the number of factors increases. It is concluded that one possibility would be to use methods which, though not fully efficient, would entail only fairly simple calculations. To devise such methods is comparatively easy, but it is far more difficult to assess their efficiency.