NAG Library Routine Document
G02BCF
1 Purpose
G02BCF computes means and standard deviations of variables, sums of squares and crossproducts of deviations from means, and Pearson productmoment correlation coefficients for a set of data omitting cases with missing values from only those calculations involving the variables for which the values are missing.
2 Specification
SUBROUTINE G02BCF ( 
N, M, X, LDX, MISS, XMISS, XBAR, STD, SSP, LDSSP, R, LDR, NCASES, CNT, LDCNT, IFAIL) 
INTEGER 
N, M, LDX, MISS(M), LDSSP, LDR, NCASES, LDCNT, IFAIL 
REAL (KIND=nag_wp) 
X(LDX,M), XMISS(M), XBAR(M), STD(M), SSP(LDSSP,M), R(LDR,M), CNT(LDCNT,M) 

3 Description
The input data consist of
$n$ observations for each of
$m$ variables, given as an array
where
${x}_{ij}$ is the
$i$th observation on the
$j$th variable. 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
$j$th variable is denoted by
${\mathit{xm}}_{j}$. Missing values need not be specified for all variables.
Let
${w}_{\mathit{i}\mathit{j}}=0$ if the
$\mathit{i}$th observation for the
$\mathit{j}$th variable is a missing value, i.e., if a missing value,
${\mathit{xm}}_{\mathit{j}}$, has been declared for the
$\mathit{j}$th variable, and
${x}_{\mathit{i}\mathit{j}}={\mathit{xm}}_{\mathit{j}}$ (see also
Section 7); and
${w}_{\mathit{i}\mathit{j}}=1$ otherwise, for
$\mathit{i}=1,2,\dots ,n$ and
$\mathit{j}=1,2,\dots ,m$.
The quantities calculated are:
(a) 
Means:

(b) 
Standard deviations:

(c) 
Sums of squares and crossproducts of deviations from means:
where
(i.e., the means used in the calculation of the sums of squares and crossproducts of deviations are based on the same set of observations as are the crossproducts.) 
(d) 
Pearson productmoment correlation coefficients:
where ${S}_{jj\left(k\right)}={\displaystyle \sum _{i=1}^{n}}{w}_{ij}{w}_{ik}{\left({x}_{ij}{\stackrel{}{x}}_{j\left(k\right)}\right)}^{2}$ and ${S}_{kk\left(j\right)}={\displaystyle \sum _{i=1}^{n}}{w}_{ik}{w}_{ij}{\left({x}_{ik}{\stackrel{}{x}}_{k\left(j\right)}\right)}^{2}$ and ${\stackrel{}{x}}_{j\left(k\right)}$ and ${\stackrel{}{x}}_{k\left(j\right)}$ are as defined in (c) above
(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 ${S}_{jj\left(k\right)}$ or ${S}_{kk\left(j\right)}$ is zero, ${R}_{jk}$ is set to zero. 
(e) 
The number of cases used in the calculation of each of the correlation coefficients:
(The diagonal terms, ${c}_{\mathit{j}\mathit{j}}$, for $\mathit{j}=1,2,\dots ,m$, also give the number of cases used in the calculation of the means, ${\stackrel{}{x}}_{\mathit{j}}$, and the standard deviations, ${s}_{\mathit{j}}$.) 
4 References
None.
5 Arguments
 1: $\mathrm{N}$ – INTEGERInput

On entry: $n$, the number of observations or cases.
Constraint:
${\mathbf{N}}\ge 2$.
 2: $\mathrm{M}$ – INTEGERInput

On entry: $m$, the number of variables.
Constraint:
${\mathbf{M}}\ge 2$.
 3: $\mathrm{X}\left({\mathbf{LDX}},{\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry: ${\mathbf{X}}\left(\mathit{i},\mathit{j}\right)$ must be set to ${x}_{\mathit{i}\mathit{j}}$, the value of the $\mathit{i}$th observation on the $\mathit{j}$th variable, for $\mathit{i}=1,2,\dots ,n$ and $\mathit{j}=1,2,\dots ,m$.
 4: $\mathrm{LDX}$ – INTEGERInput

On entry: the first dimension of the array
X as declared in the (sub)program from which G02BCF is called.
Constraint:
${\mathbf{LDX}}\ge {\mathbf{N}}$.
 5: $\mathrm{MISS}\left({\mathbf{M}}\right)$ – INTEGER arrayInput

On entry:
${\mathbf{MISS}}\left(j\right)$ must be set equal to
$1$ if a missing value,
$x{m}_{j}$, is to be specified for the
$j$th 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.
 6: $\mathrm{XMISS}\left({\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayInput

On entry:
${\mathbf{XMISS}}\left(j\right)$ must be set to the missing value,
$x{m}_{j}$, to be associated with the
$j$th variable in the array
X, for those variables for which missing values are specified by means of the array
MISS (see
Section 7).
 7: $\mathrm{XBAR}\left({\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: the mean value,
${\stackrel{}{x}}_{\mathit{j}}$, of the $\mathit{j}$th variable, for $\mathit{j}=1,2,\dots ,m$.
 8: $\mathrm{STD}\left({\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: the standard deviation,
${s}_{\mathit{j}}$, of the $\mathit{j}$th variable, for $\mathit{j}=1,2,\dots ,m$.
 9: $\mathrm{SSP}\left({\mathbf{LDSSP}},{\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{SSP}}\left(\mathit{j},\mathit{k}\right)$ is the crossproduct of deviations ${S}_{\mathit{j}\mathit{k}}$, for $\mathit{j}=1,2,\dots ,m$ and $\mathit{k}=1,2,\dots ,m$.
 10: $\mathrm{LDSSP}$ – INTEGERInput

On entry: the first dimension of the array
SSP as declared in the (sub)program from which G02BCF is called.
Constraint:
${\mathbf{LDSSP}}\ge {\mathbf{M}}$.
 11: $\mathrm{R}\left({\mathbf{LDR}},{\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{R}}\left(\mathit{j},\mathit{k}\right)$ is the productmoment correlation coefficient ${R}_{\mathit{j}\mathit{k}}$ between the $\mathit{j}$th and $\mathit{k}$th variables, for $\mathit{j}=1,2,\dots ,m$ and $\mathit{k}=1,2,\dots ,m$.
 12: $\mathrm{LDR}$ – INTEGERInput

On entry: the first dimension of the array
R as declared in the (sub)program from which G02BCF is called.
Constraint:
${\mathbf{LDR}}\ge {\mathbf{M}}$.
 13: $\mathrm{NCASES}$ – INTEGEROutput

On exit: the minimum number of cases used in the calculation of any of the sums of squares and crossproducts and correlation coefficients (when cases involving missing values have been eliminated).
 14: $\mathrm{CNT}\left({\mathbf{LDCNT}},{\mathbf{M}}\right)$ – REAL (KIND=nag_wp) arrayOutput

On exit: ${\mathbf{CNT}}\left(\mathit{j},\mathit{k}\right)$ is the number of cases, ${c}_{\mathit{j}\mathit{k}}$, actually used in the calculation of ${S}_{\mathit{j}\mathit{k}}$, and ${R}_{\mathit{j}\mathit{k}}$, the sum of crossproducts and correlation coefficient for the $\mathit{j}$th and $\mathit{k}$th variables, for $\mathit{j}=1,2,\dots ,m$ and $\mathit{k}=1,2,\dots ,m$.
 15: $\mathrm{LDCNT}$ – INTEGERInput

On entry: the first dimension of the array
CNT as declared in the (sub)program from which G02BCF is called.
Constraint:
${\mathbf{LDCNT}}\ge {\mathbf{M}}$.
 16: $\mathrm{IFAIL}$ – INTEGERInput/Output

On entry:
IFAIL must be set to
$0$,
$1\text{ or}1$. If you are unfamiliar with this argument you should refer to
Section 3.4 in How to Use the NAG Library and its Documentation for details.
For environments where it might be inappropriate to halt program execution when an error is detected, the value
$1\text{ or}1$ is recommended. If the output of error messages is undesirable, then the value
$1$ is recommended. Otherwise, because for this routine the values of the output arguments may be useful even if
${\mathbf{IFAIL}}\ne {\mathbf{0}}$ on exit, the recommended value is
$1$.
When the value $\mathbf{1}\text{ or}1$ is used it is essential to test the value of IFAIL on exit.
On exit:
${\mathbf{IFAIL}}={\mathbf{0}}$ unless the routine detects an error or a warning has been flagged (see
Section 6).
6 Error Indicators and Warnings
If on entry
${\mathbf{IFAIL}}=0$ or
$1$, explanatory error messages are output on the current error message unit (as defined by
X04AAF).
Note: G02BCF may return useful information for one or more of the following detected errors or warnings.
Errors or warnings detected by the routine:
 ${\mathbf{IFAIL}}=1$

On entry,  ${\mathbf{N}}<2$. 
 ${\mathbf{IFAIL}}=2$

On entry,  ${\mathbf{M}}<2$. 
 ${\mathbf{IFAIL}}=3$

On entry,  ${\mathbf{LDX}}<{\mathbf{N}}$, 
or  ${\mathbf{LDSSP}}<{\mathbf{M}}$, 
or  ${\mathbf{LDR}}<{\mathbf{M}}$, 
or  ${\mathbf{LDCNT}}<{\mathbf{M}}$. 
 ${\mathbf{IFAIL}}=4$

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 crossproducts, and correlation coefficients based on two or more cases are returned by the routine even if
${\mathbf{IFAIL}}={\mathbf{4}}$.
 ${\mathbf{IFAIL}}=99$
An unexpected error has been triggered by this routine. Please
contact
NAG.
See
Section 3.9 in How to Use the NAG Library and its Documentation for further information.
 ${\mathbf{IFAIL}}=399$
Your licence key may have expired or may not have been installed correctly.
See
Section 3.8 in How to Use the NAG Library and its Documentation for further information.
 ${\mathbf{IFAIL}}=999$
Dynamic memory allocation failed.
See
Section 3.7 in How to Use the NAG Library and its Documentation for further information.
7 Accuracy
G02BCF 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. G02BCF treats all values in the inclusive range
$\left(1\pm {0.1}^{\left({\mathbf{X02BEF}}2\right)}\right)\times {xm}_{j}$, where
${\mathit{xm}}_{j}$ 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.
8 Parallelism and Performance
G02BCF is not threaded in any implementation.
The time taken by G02BCF depends on $n$ and $m$, and the occurrence of missing values.
The routine uses a twopass algorithm.
10 Example
This example reads in a set of data consisting of five observations on each of three variables. Missing values of $0.0$, $1.0$ and $0.0$ are declared for the first, second and third variables respectively. The means, standard deviations, sums of squares and crossproducts of deviations from means, and Pearson productmoment correlation coefficients for all three 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 omits cases $4$ and $5$ in calculating the correlation between the first and second variables, and cases $3$ and $4$ for the first and third variables etc.
10.1 Program Text
Program Text (g02bcfe.f90)
10.2 Program Data
Program Data (g02bcfe.d)
10.3 Program Results
Program Results (g02bcfe.r)