S3methodsFAiR {FAiR}R Documentation

S3 methods for objects of class "FA"

Description

These S3 methods for objects of class "FA" or that inherit from class "FA" provide fairly standard post-estimation functions for factor analysis models.

Usage

## S3 method for class 'FA':
deviance(object, ...)
## S3 method for class 'FA.general':
deviance(object, ...)
## S3 method for class 'FA.2ndorder':
deviance(object, ...)
## S3 method for class 'FA':
df.residual(object, ...)
## S3 method for class 'FA.general':
df.residual(object, ...)
## S3 method for class 'FA.2ndorder':
df.residual(object, ...)
## S3 method for class 'FA':
fitted(object, ...)
## S3 method for class 'FA.general':
fitted(object, ...)
## S3 method for class 'FA.2ndorder':
fitted(object, ...)
## S3 method for class 'FA':
influence(model, ...)
## S3 method for class 'FA.general':
influence(model, ...)
## S3 method for class 'FA.2ndorder':
influence(model, ...)
## S3 method for class 'FA':
model.matrix(object, ...)
## S3 method for class 'FA.general':
model.matrix(object, ...)
## S3 method for class 'FA.2ndorder':
model.matrix(object, ...)
## S3 method for class 'FA':
pairs(x, ...)
## S3 method for class 'FA.general':
pairs(x, ...)
## S3 method for class 'FA.2ndorder':
pairs(x, ...)
## S3 method for class 'FA':
predict(object, ...)
## S3 method for class 'FA.general':
predict(object, ...)
## S3 method for class 'FA.2ndorder':
predict(object, ...)
## S3 method for class 'FA':
residuals(object, ...)
## S3 method for class 'FA.general':
residuals(object, ...)
## S3 method for class 'FA.2ndorder':
residuals(object, ...)
## S3 method for class 'FA':
rstandard(model, ...)
## S3 method for class 'FA.general':
rstandard(model, ...)
## S3 method for class 'FA.2ndorder':
rstandard(model, ...)
## S3 method for class 'FA':
weights(object, ...)
## S3 method for class 'FA.general':
weights(object, ...)
## S3 method for class 'FA.2ndorder':
weights(object, ...)

Arguments

object An object of class 'FA' or that inherits from 'FA'
model An object of class 'FA' or that inherits from 'FA'
x An object of class 'FA' or that inherits from 'FA'
... additional argument(s) for methods

Details

The code for all of these methods is quite short. Feel free to simply call body() on the method.FA to see what the function does. There is no difference in functionality for the methods that inherit from class 'FA' relative to those that are defined for class 'FA'.

Value

deviance returns a scalar indicating the discrepancy from a perfect fitting model.
df.residual returns an integer indicating the difference between the number of nonredundant elements in the sample correlation matrix and the number of estimated parameters.
fitted returns a square matrix of estimated correlations in the common factor space with communalities along the diagonal.
influence returns a square matrix that is equal to residuals() * weights().
model.matrix returns the sample correlation matrix among outcomes.
pairs returns nothing but plots the estimated reference structure correlations; to repeat: these are the reference structure correlations, rather than the primary pattern coefficients.
predict returns a matrix with the predicted values of the outcomes, which is defined as the product of the factor score matrix and the primary pattern coefficient matrix. Thus, an error is returned if factor scores were not calculated by Factanal.
residuals returns a square matrix that contains the difference between model.matrix() and fitted() and thus has uniquenesses along the diagonal.
rstandard returns residuals rescaled into a correlation matrix and thus has ones along the diagonal.
weights returns a square matrix with the weights used in the discrepancy function. For Yates' weighted least squares estimator these weights are as defined in equation 188. For maximum likelihood estimation, these weights are proportional to the reciprocal of the crossproduct of the uniquenesses and are only approximately equal to the implied weights that would be used if minimizing the weighted sum of squared residuals. For ease of interpretation they are rescaled so that the mean weight is 1.0.

Author(s)

Ben Goodrich http://wiki.r-project.org/rwiki/doku.php?id=packages:cran:fair

References

Yates, A. (1987) Multivariate Exploratory Data Analysis: A Perspective on Exploratory Factor Analysis. State University of New York Press.

See Also

confint, deviance, df.residual, fitted, influence, model.matrix, pairs, predict, residuals, rstandard, FA-class.

Examples

  ## See the example for Factanal()

[Package FAiR version 0.2-0 Index]