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:
factanal , predict .