NAG FL Interface
g13dnf (multi_​corrmat_​partlag)

1 Purpose

g13dnf calculates the sample partial lag correlation matrices of a multivariate time series. A set of χ2-statistics and their significance levels are also returned. A call to g13dmf is usually made prior to calling this routine in order to calculate the sample cross-correlation matrices.

2 Specification

Fortran Interface
Subroutine g13dnf ( k, n, m, kmax, r0, r, maxlag, parlag, x, pvalue, work, lwork, ifail)
Integer, Intent (In) :: k, n, m, kmax, lwork
Integer, Intent (Inout) :: ifail
Integer, Intent (Out) :: maxlag
Real (Kind=nag_wp), Intent (In) :: r0(kmax,k), r(kmax,kmax,m)
Real (Kind=nag_wp), Intent (Inout) :: parlag(kmax,kmax,m)
Real (Kind=nag_wp), Intent (Out) :: x(m), pvalue(m), work(lwork)
C Header Interface
#include <nag.h>
void  g13dnf_ (const Integer *k, const Integer *n, const Integer *m, const Integer *kmax, const double r0[], const double r[], Integer *maxlag, double parlag[], double x[], double pvalue[], double work[], const Integer *lwork, Integer *ifail)
The routine may be called by the names g13dnf or nagf_tsa_multi_corrmat_partlag.

3 Description

Let Wt = w1t,w2t,,wkt T , for t=1,2,,n, denote n observations of a vector of k time series. The partial lag correlation matrix at lag l, Pl, is defined to be the correlation matrix between Wt and Wt+l, after removing the linear dependence on each of the intervening vectors Wt+1,Wt+2,,Wt+l-1. It is the correlation matrix between the residual vectors resulting from the regression of Wt+l on the carriers Wt+l-1,,Wt+1 and the regression of Wt on the same set of carriers; see Heyse and Wei (1985).
Pl has the following properties.
  1. (i)If Wt follows a vector autoregressive model of order p, then Pl=0 for l>p;
  2. (ii)When k=1, Pl reduces to the univariate partial autocorrelation at lag l;
  3. (iii)Each element of Pl is a properly normalized correlation coefficient;
  4. (iv)When l=1, Pl is equal to the cross-correlation matrix at lag 1 (a natural property which also holds for the univariate partial autocorrelation function).
Sample estimates of the partial lag correlation matrices may be obtained using the recursive algorithm described in Wei (1990). They are calculated up to lag m, which is usually taken to be at most n/4. Only the sample cross-correlation matrices (R^l, for l=0,1,,m) and the standard deviations of the series are required as input to g13dnf. These may be computed by g13dmf. Under the hypothesis that Wt follows an autoregressive model of order s-1, the elements of the sample partial lag matrix P^s, denoted by P^ijs, are asymptotically Normally distributed with mean zero and variance 1/n. In addition the statistic
Xs=ni=1kj=1kP^ij s 2  
has an asymptotic χ2-distribution with k2 degrees of freedom. These quantities, Xl, are useful as a diagnostic aid for determining whether the series follows an autoregressive model and, if so, of what order.

4 References

Heyse J F and Wei W W S (1985) The partial lag autocorrelation function Technical Report No. 32 Department of Statistics, Temple University, Philadelphia
Wei W W S (1990) Time Series Analysis: Univariate and Multivariate Methods Addison–Wesley

5 Arguments

1: k Integer Input
On entry: k, the dimension of the multivariate time series.
Constraint: k1.
2: n Integer Input
On entry: n, the number of observations in each series.
Constraint: n2.
3: m Integer Input
On entry: m, the number of partial lag correlation matrices to be computed. Note this also specifies the number of sample cross-correlation matrices that must be contained in the array r.
Constraint: 1m<n.
4: kmax Integer Input
On entry: the first dimension of the arrays r0, r and parlag and the second dimension of the arrays r and parlag as declared in the (sub)program from which g13dnf is called.
Constraint: kmaxk.
5: r0kmaxk Real (Kind=nag_wp) array Input
On entry: if ij, then r0ij must contain the i,jth element of the sample cross-correlation matrix at lag zero, R^ij0. If i=j, then r0ii must contain the standard deviation of the ith series.
6: rkmaxkmaxm Real (Kind=nag_wp) array Input
On entry: rijl must contain the i,jth element of the sample cross-correlation at lag l, R^ijl, for l=1,2,,m, i=1,2,,k and j=1,2,,k, where series j leads series i (see Section 9).
7: maxlag Integer Output
On exit: the maximum lag up to which partial lag correlation matrices (along with χ2-statistics and their significance levels) have been successfully computed. On a successful exit maxlag will equal m. If ifail=2 on exit, maxlag will be less than m.
8: parlagkmaxkmaxm Real (Kind=nag_wp) array Output
On exit: parlagijl contains the i,jth element of the sample partial lag correlation matrix at lag l, P^ijl, for l=1,2,,maxlag, i=1,2,,k and j=1,2,,k.
9: xm Real (Kind=nag_wp) array Output
On exit: xl contains the χ2-statistic at lag l, for l=1,2,,maxlag.
10: pvaluem Real (Kind=nag_wp) array Output
On exit: pvaluel contains the significance level of the corresponding χ2-statistic in x, for l=1,2,,maxlag.
11: worklwork Real (Kind=nag_wp) array Workspace
12: lwork Integer Input
On entry: the dimension of the array work as declared in the (sub)program from which g13dnf is called.
Constraint: lwork5m+6k2+k.
13: ifail Integer Input/Output
On entry: ifail must be set to 0, -1 or 1 to set behaviour on detection of an error; these values have no effect when no error is detected.
A value of 0 causes the printing of an error message and program execution will be halted; otherwise program execution continues. A value of -1 means that an error message is printed while a value of 1 means that it is not.
If halting is not appropriate, the value -1 or 1 is recommended. If message printing is undesirable, then the value 1 is recommended. Otherwise, the value 0 is recommended. When the value -1 or 1 is used it is essential to test the value of ifail on exit.
On exit: ifail=0 unless the routine detects an error or a warning has been flagged (see Section 6).

6 Error Indicators and Warnings

If on entry ifail=0 or -1, explanatory error messages are output on the current error message unit (as defined by x04aaf).
Errors or warnings detected by the routine:
ifail=1
On entry, k=value.
Constraint: k1.
On entry, kmax=value and k=value.
Constraint: kmaxk.
On entry, lwork=value and the minimum size required=value.
Constraint: lwork5×m+6×k2+k.
On entry, m=value.
Constraint: m1.
On entry, m=value and n=value.
Constraint: m<n.
On entry, n=value.
Constraint: n2.
ifail=2
The recursive equations used to compute the partial lag correlation matrices are ill-conditioned (they have been computed up to lag value). All output quantities in the arrays parlag, x and pvalue up to and including lag maxlag will be correct.
ifail=-99
An unexpected error has been triggered by this routine. Please contact NAG.
See Section 7 in the Introduction to the NAG Library FL Interface for further information.
ifail=-399
Your licence key may have expired or may not have been installed correctly.
See Section 8 in the Introduction to the NAG Library FL Interface for further information.
ifail=-999
Dynamic memory allocation failed.
See Section 9 in the Introduction to the NAG Library FL Interface for further information.

7 Accuracy

The accuracy will depend upon the accuracy of the sample cross-correlations.

8 Parallelism and Performance

g13dnf is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
g13dnf makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the X06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this routine. Please also consult the Users' Note for your implementation for any additional implementation-specific information.

9 Further Comments

The time taken is roughly proportional to m2k3.
If you have calculated the sample cross-correlation matrices in the arrays r0 and r, without calling g13dmf, then care must be taken to ensure they are supplied as described in Section 5. In particular, for l1, R^ijl must contain the sample cross-correlation coefficient between wit-l and wjt.
The routine g13dbf computes squared partial autocorrelations for a specified number of lags. It may also be used to estimate a sequence of partial autoregression matrices at lags 1,2, by making repeated calls to the routine with the argument nk set to 1,2,. The i,jth element of the sample partial autoregression matrix at lag l is given by Wi,j,l when nk is set equal to l on entry to g13dbf. Note that this is the ‘Yule–Walker’ estimate. Unlike the partial lag correlation matrices computed by g13dnf, when Wt follows an autoregressive model of order s-1, the elements of the sample partial autoregressive matrix at lag s do not have variance 1/n, making it very difficult to spot a possible cut-off point. The differences between these matrices are discussed further by Wei (1990).
Note that g13dbf takes the sample cross-covariance matrices as input whereas this routine requires the sample cross-correlation matrices to be input.

10 Example

This example computes the sample partial lag correlation matrices of two time series of length 48, up to lag 10. The matrices, their χ2-statistics and significance levels and a plot of symbols indicating which elements of the sample partial lag correlation matrices are significant are printed. Three * represent significance at the 0.5% level, two * represent significance at the 1% level and a single * represents significance at the 5% level. The * are plotted above or below the central line depending on whether the elements are significant in a positive or negative direction.

10.1 Program Text

Program Text (g13dnfe.f90)

10.2 Program Data

Program Data (g13dnfe.d)

10.3 Program Results

Program Results (g13dnfe.r)