nag_mv_orthomax (g03bac) (PDF version)
g03 Chapter Contents
g03 Chapter Introduction
NAG Library Manual

NAG Library Function Document

nag_mv_orthomax (g03bac)

+ Contents

    1  Purpose
    7  Accuracy

1  Purpose

nag_mv_orthomax (g03bac) computes orthogonal rotations for a matrix of loadings using a generalized orthomax criterion.

2  Specification

#include <nag.h>
#include <nagg03.h>
void  nag_mv_orthomax (Nag_RotationLoading stand, double g, Integer nvar, Integer k, double fl[], Integer tdf, double flr[], double r[], Integer tdr, double acc, Integer maxit, Integer *iter, NagError *fail)

3  Description

Let Λ  be the p  by k  matrix of loadings from a variable-directed multivariate method, e.g., canonical variate analysis or factor analysis. This matrix represents the relationship between the original p  variables and the k  orthogonal linear combinations of these variables, the canonical variates or factors. The latter are only unique up to a rotation in the k -dimensional space they define. A rotation can then be found that simplifies the structure of the matrix of loadings, and hence the relationship between the original and the derived variables. That is, the elements, λ ij * , of the rotated matrix, Λ * , are either relatively large or small. The rotations may be found by minimizing the criterion:
V = j=1 k i=1 p λ ij * 4 - γ p j=1 k i=1 p λ ij * 2 2
where the constant γ  gives a family of rotations with γ=1  giving varimax rotations and γ=0  giving quartimax rotations.
It is generally advised that factor loadings should be standardized, so that the sum of squared elements for each row is one, before computing the rotations.
The matrix of rotations, R , such that Λ * = Λ R , is computed using first an algorithm based on that described by Cooley and Lohnes (1971), which involves the pairwise rotation of the factors. Then a final refinement is made using a method similar to that described by Lawley and Maxwell (1971), but instead of the eigenvalue decomposition, the algorithm has been adapted to incorporate a singular value decomposition.

4  References

Cooley W C and Lohnes P R (1971) Multivariate Data Analysis Wiley
Lawley D N and Maxwell A E (1971) Factor Analysis as a Statistical Method (2nd Edition) Butterworths

5  Arguments

1:     standNag_RotationLoadingInput
On entry: indicates if the matrix of loadings is to be row standardized before rotation.
stand=Nag_RoLoadStand
The loadings are row standardized.
stand=Nag_RoLoadNotStand
The loadings are left unstandardized.
Constraint: stand=Nag_RoLoadStand or Nag_RoLoadNotStand.
2:     gdoubleInput
On entry: the criterion constant, γ , with γ=1.0  giving varimax rotations and γ=0.0  giving quartimax rotations.
Constraint: g0.0 .
3:     nvarIntegerInput
On entry: the number of original variables, p .
Constraint: nvark .
4:     kIntegerInput
On entry: the number of derived variates or factors, k .
Constraint: k2 .
5:     fl[nvar×tdf]doubleInput/Output
On entry: the matrix of loadings, Λ . fl[i-1×tdf+j-1]  must contain the loading for the i th variable on the j th factor, for i=1,2,,p and j=1,2,,k.
On exit: if stand=Nag_RoLoadStand the elements of fl are standardized so that the sum of squared elements for each row is 1.0 and then after, the computation of the rotations are rescaled; this may lead to slight differences between the input and output values of fl. If stand=Nag_RoLoadNotStand, fl will be unchanged on exit.
6:     tdfIntegerInput
On entry: the stride separating matrix column elements in the arrays fl, flr.
Constraint: tdfk .
7:     flr[nvar×tdf]doubleOutput
On exit: the rotated matrix of loadings, Λ * . flr[i-1×tdf+j-1]  will contain the rotated loading for the i th variable on the j th factor, for i=1,2,,p and j=1,2,,k.
8:     r[k×tdr]doubleOutput
Note: the i,jth element of the matrix R is stored in r[i-1×tdr+j-1].
On exit: the matrix of rotations, R .
9:     tdrIntegerInput
On entry: the stride separating matrix column elements in the array r.
Constraint: tdrk .
10:   accdoubleInput
On entry: indicates the accuracy required. The iterative procedure of Cooley and Lohnes (1971) will be stopped and the final refinement computed when the change in V  is less than acc × max1.0,V . If acc is greater than or equal to 0.0 but less than machine precision, or if acc is greater than 1.0, then machine precision will be used instead.
It is suggested that acc be set to 0.00001.
Constraint: acc0.0 .
11:   maxitIntegerInput
On entry: the maximum number of iterations. It is suggested that maxit be set to 30.
Constraint: maxit1 .
12:   iterInteger *Output
On exit: the number of iterations performed.
13:   failNagError *Input/Output
The NAG error argument (see Section 3.6 in the Essential Introduction).

6  Error Indicators and Warnings

NE_2_INT_ARG_LT
On entry, nvar=value  while k=value . These arguments must satisfy nvark .
On entry, tdf=value  while k=value . These arguments must satisfy tdfk .
On entry, tdr=value  while k=value . These arguments must satisfy tdrk .
NE_ACC_ITER
The algorithm to find R has failed to reach the required accuracy in the given number of iterations, value. Try increasing acc or increasing maxit. The returned solution should be a reasonable approximation.
NE_ALLOC_FAIL
Dynamic memory allocation failed.
NE_BAD_PARAM
On entry, argument stand had an illegal value.
NE_INT_ARG_LE
On entry, maxit=value.
Constraint: maxit>0.
NE_INT_ARG_LT
On entry, k=value.
Constraint: k2.
NE_INTERNAL_ERROR
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
NE_REAL_ARG_LT
On entry, acc must not be less than 0.0 : acc=value .
On entry, g must not be less than 0.0 : g=value .
NE_SVD_NOT_CONV
The singular value decomposition has failed to converge. This is an unlikely error exit.

7  Accuracy

The accuracy is determined by the value of acc.

8  Parallelism and Performance

Not applicable.

9  Further Comments

If the results of a principal component analysis as carried out by nag_mv_prin_comp (g03aac) are to be rotated, the loadings as returned in the array p  by nag_mv_prin_comp (g03aac) can be supplied via the argument fl to nag_mv_orthomax (g03bac). The resulting rotation matrix can then be used to rotate the principal component scores as returned in the array v  by nag_mv_prin_comp (g03aac). The function nag_dgemm (f16yac) may be used for this matrix multiplication.

10  Example

The example is taken from page 75 of Lawley and Maxwell (1971). The results from a factor analysis of ten variables using three factors are input and rotated using varimax rotations without standardizing rows.

10.1  Program Text

Program Text (g03bace.c)

10.2  Program Data

Program Data (g03bace.d)

10.3  Program Results

Program Results (g03bace.r)


nag_mv_orthomax (g03bac) (PDF version)
g03 Chapter Contents
g03 Chapter Introduction
NAG Library Manual

© The Numerical Algorithms Group Ltd, Oxford, UK. 2014