Factor Analysis Scores

DESCRIPTION:

Produces a matrix of factor scores for the model.

USAGE:

predict.factanal(object, newdata=NULL, type="regression") 

REQUIRED ARGUMENTS:

object
an object of class "factanal".

OPTIONAL ARGUMENTS:

newdata
a matrix or data frame of data containing variables with the same names as those used in object.
type
the estimation type for the scores. Choices are "regression", "weighted.ls" or "least.squares". Only the first letter is necessary.

VALUE:

a matrix of scores with an attribute "type" giving the type of score computed.

DETAILS:

The regression type of score minimizes the mean squared error. The weighted least squares type (due to Bartlett) is unbiased but has a larger variance.

REFERENCES:

Johnson, R. A. and Wichern, D. W. (1982). Applied Multivariate Statistical Analysis. Prentice-Hall, Englewood Cliffs, New Jersey.

Seber, G. A. F. (1984). Multivariate Observations. Wiley, New York.

SEE ALSO:

, .