g02bj computes means and standard deviations, sums of squares and cross-products of deviations from means, and Pearson product-moment correlation coefficients for selected variables omitting cases with missing values from only those calculations involving the variables for which the values are missing.

Syntax

C#
public static void g02bj(
	int n,
	int m,
	double[,] x,
	int[] miss,
	double[] xmiss,
	int nvars,
	int[] kvar,
	double[] xbar,
	double[] std,
	double[,] ssp,
	double[,] r,
	out int ncases,
	double[,] cnt,
	out int ifail
)
Visual Basic
Public Shared Sub g02bj ( _
	n As Integer, _
	m As Integer, _
	x As Double(,), _
	miss As Integer(), _
	xmiss As Double(), _
	nvars As Integer, _
	kvar As Integer(), _
	xbar As Double(), _
	std As Double(), _
	ssp As Double(,), _
	r As Double(,), _
	<OutAttribute> ByRef ncases As Integer, _
	cnt As Double(,), _
	<OutAttribute> ByRef ifail As Integer _
)
Visual C++
public:
static void g02bj(
	int n, 
	int m, 
	array<double,2>^ x, 
	array<int>^ miss, 
	array<double>^ xmiss, 
	int nvars, 
	array<int>^ kvar, 
	array<double>^ xbar, 
	array<double>^ std, 
	array<double,2>^ ssp, 
	array<double,2>^ r, 
	[OutAttribute] int% ncases, 
	array<double,2>^ cnt, 
	[OutAttribute] int% ifail
)
F#
static member g02bj : 
        n : int * 
        m : int * 
        x : float[,] * 
        miss : int[] * 
        xmiss : float[] * 
        nvars : int * 
        kvar : int[] * 
        xbar : float[] * 
        std : float[] * 
        ssp : float[,] * 
        r : float[,] * 
        ncases : int byref * 
        cnt : float[,] * 
        ifail : int byref -> unit 

Parameters

n
Type: System..::..Int32
On entry: n, the number of observations or cases.
Constraint: n2.
m
Type: System..::..Int32
On entry: m, the number of variables.
Constraint: m2.
x
Type: array<System..::..Double,2>[,](,)[,][,]
An array of size [dim1, m]
Note: dim1 must satisfy the constraint: dim1n
On entry: x[i-1,j-1] must be set to xij, the value of the ith observation on the jth variable, for i=1,2,,n and j=1,2,,m.
miss
Type: array<System..::..Int32>[]()[][]
An array of size [m]
On entry: miss[j-1] must be set equal to 1 if a missing value, xmj, is to be specified for the jth variable in the array x, or set equal to 0 otherwise. Values of miss must be given for all m variables in the array x.
xmiss
Type: array<System..::..Double>[]()[][]
An array of size [m]
On entry: xmiss[j-1] must be set to the missing value, xmj, to be associated with the jth variable in the array x, for those variables for which missing values are specified by means of the array miss (see [Accuracy]).
nvars
Type: System..::..Int32
On entry: p, the number of variables for which information is required.
Constraint: 2nvarsm.
kvar
Type: array<System..::..Int32>[]()[][]
An array of size [nvars]
On entry: kvar[j] must be set to the column number in x of the jth variable for which information is required, for j=0,1,,p-1.
Constraint: 1kvar[j]m, for j=0,1,,p-1.
xbar
Type: array<System..::..Double>[]()[][]
An array of size [nvars]
On exit: the mean value, x-j, of the variable specified in kvar[j-1], for j=1,2,,p.
std
Type: array<System..::..Double>[]()[][]
An array of size [nvars]
On exit: the standard deviation, sj, of the variable specified in kvar[j-1], for j=1,2,,p.
ssp
Type: array<System..::..Double,2>[,](,)[,][,]
An array of size [dim1, nvars]
Note: dim1 must satisfy the constraint: dim1nvars
On exit: ssp[j-1,k-1] is the cross-product of deviations, Sjk, for the variables specified in kvar[j-1] and kvar[k-1], for j=1,2,,p and k=1,2,,p.
r
Type: array<System..::..Double,2>[,](,)[,][,]
An array of size [dim1, nvars]
Note: dim1 must satisfy the constraint: dim1nvars
On exit: r[j-1,k-1] is the product-moment correlation coefficient, Rjk, between the variables specified in kvar[j-1] and kvar[k-1], for j=1,2,,p and k=1,2,,p.
ncases
Type: System..::..Int32%
On exit: the minimum number of cases used in the calculation of any of the sums of squares and cross-products and correlation coefficients (when cases involving missing values have been eliminated).
cnt
Type: array<System..::..Double,2>[,](,)[,][,]
An array of size [dim1, nvars]
Note: dim1 must satisfy the constraint: dim1nvars
On exit: cnt[j-1,k-1] is the number of cases, cjk, actually used in the calculation of Sjk, and Rjk, the sum of cross-products and correlation coefficient for the variables specified in kvar[j-1] and kvar[k-1], for j=1,2,,p and k=1,2,,p.
ifail
Type: System..::..Int32%
On exit: ifail=0 unless the method detects an error or a warning has been flagged (see [Error Indicators and Warnings]).

Description

The input data consists of n observations for each of m variables, given as an array
xij,  i=1,2,,nn2,j=1,2,,mm2,
where xij is the ith observation on the jth variable, together with the subset of these variables, v1,v2,,vp, for which information is required.
In addition, each of the m variables may optionally have associated with it a value which is to be considered as representing a missing observation for that variable; the missing value for the jth variable is denoted by xmj. Missing values need not be specified for all variables.
Let wij=0 if the ith observation for the jth variable is a missing value, i.e., if a missing value, xmj, has been declared for the jth variable, and xij=xmj (see also [Accuracy]); and wij=1 otherwise, for i=1,2,,n and j=1,2,,m.
The quantities calculated are:
(a) Means:
x-j=i=1nwijxiji=1nwij,  j=v1,v2,,vp.
(b) Standard deviations:
sj=i=1nwijxij-x-j2i=1nwij-1,  j=v1,v2,,vp.
(c) Sums of squares and cross-products of deviations from means:
Sjk=i=1nwijwikxij-x-jkxik-x-kj,  j,k=v1,v2,,vp,
where
x-jk=i=1nwijwikxiji=1nwijwik  and  x-kj=i=1nwikwijxiki=1nwikwij,
(i.e., the means used in the calculation of the sum of squares and cross-products of deviations are based on the same set of observations as are the cross-products).
(d) Pearson product-moment correlation coefficients:
Rjk=SjkSjjkSkkj,  j,k=v1,v2,,vp,
where
Sjjk=i=1nwijwikxij-x-jk2  and  Skkj=i=1nwikwijxik-x-kj2,
(i.e., the sums of squares of deviations used in the denominator are based on the same set of observations as are used in the calculation of the numerator).
If Sjjk or Skkj is zero, Rjk is set to zero.
(e) The number of cases used in the calculation of each of the correlation coefficients:
cjk=i=1nwijwik,  j,k=v1,v2,,vp.
(The diagonal terms, cjj, for j=v1,v2,,vp, also give the number of cases used in the calculation of the means, x-j, and the standard deviations, sj.)

References

None.

Error Indicators and Warnings

Note: g02bj may return useful information for one or more of the following detected errors or warnings.
Errors or warnings detected by the method:
Some error messages may refer to parameters that are dropped from this interface (LDX, LDSSP, LDR, LDCNT) In these cases, an error in another parameter has usually caused an incorrect value to be inferred.
ifail=1
On entry,n<2.
ifail=2
On entry,nvars<2,
ornvars>m.
ifail=4
On entry,kvar[j-1]<1,
orkvar[j-1]>m for some j=1,2,,nvars.
ifail=5
After observations with missing values were omitted, fewer than two cases remained for at least one pair of variables. (The pairs of variables involved can be determined by examination of the contents of the array cnt.) All means, standard deviations, sums of squares and cross-products, and correlation coefficients based on two or more cases are returned by the method even if ifail=5.
ifail=-9000
An error occured, see message report.
ifail=-6000
Invalid Parameters value
ifail=-4000
Invalid dimension for array value
ifail=-8000
Negative dimension for array value
ifail=-6000
Invalid Parameters value

Accuracy

g02bj does not use additional precision arithmetic for the accumulation of scalar products, so there may be a loss of significant figures for large n.
You are warned of the need to exercise extreme care in your selection of missing values. g02bj treats all values in the inclusive range 1±0.1x02be-2×xmj, where xmj is the missing value for variable j specified in xmiss.
You must therefore ensure that the missing value chosen for each variable is sufficiently different from all valid values for that variable so that none of the valid values fall within the range indicated above.

Parallelism and Performance

None.

Further Comments

The time taken by g02bj depends on n and p, and the occurrence of missing values.
The method uses a two-pass algorithm.

Example

This example reads in a set of data consisting of five observations on each of four variables. Missing values of -1.0, 0.0 and 0.0 are declared for the first, second and fourth variables respectively; no missing value is specified for the third variable. The means, standard deviations, sums of squares and cross-products of deviations from means, and Pearson product-moment correlation coefficients for the fourth, first and second variables are then calculated and printed, omitting cases with missing values from only those calculations involving the variables for which the values are missing. The program therefore eliminates cases 4 and 5 in calculating the correlation between the fourth and first variables, and cases 3 and 4 for the fourth and second variables etc.

Example program (C#): g02bje.cs

Example program data: g02bje.d

Example program results: g02bje.r

See Also