NAG Library Function Document
nag_tsa_varma_estimate (g13ddc)
1
Purpose
nag_tsa_varma_estimate (g13ddc) fits a vector autoregressive moving average (VARMA) model to an observed vector of time series using the method of Maximum Likelihood (ML). Standard errors of parameter estimates are computed along with their appropriate correlation matrix. The function also calculates estimates of the residual series.
2
Specification
#include <nag.h> |
#include <nagg13.h> |
void |
nag_tsa_varma_estimate (Integer k,
Integer n,
Integer ip,
Integer iq,
Nag_IncludeMean mean,
double par[],
Integer npar,
double qq[],
Integer kmax,
const double w[],
const Nag_Boolean parhld[],
Nag_Boolean exact,
Integer iprint,
double cgetol,
Integer maxcal,
Integer ishow,
const char *outfile,
Integer *niter,
double *rlogl,
double v[],
double g[],
double cm[],
Integer pdcm,
NagError *fail) |
|
3
Description
Let
, for
, denote a vector of
time series which is assumed to follow a multivariate ARMA model of the form
where
, for
, is a vector of
residual series assumed to be Normally distributed with zero mean and positive definite covariance matrix
. The components of
are assumed to be uncorrelated at non-simultaneous lags. The
and
are
by
matrices of parameters.
, for
, are called the autoregressive (AR) parameter matrices, and
, for
, the moving average (MA) parameter matrices. The parameters in the model are thus the
(
by
)
-matrices, the
(
by
)
-matrices, the mean vector,
, and the residual error covariance matrix
. Let
where
denotes the
by
identity matrix.
The ARMA model
(1) is said to be stationary if the eigenvalues of
lie inside the unit circle. Similarly, the ARMA model
(1) is said to be invertible if the eigenvalues of
lie inside the unit circle.
The method of computing the exact likelihood function (using a Kalman filter algorithm) is discussed in
Shea (1987). A quasi-Newton algorithm (see
Gill and Murray (1972)) is then used to search for the maximum of the log-likelihood function. Stationarity and invertibility are enforced on the model using the reparameterisation discussed in
Ansley and Kohn (1986). Conditional on the maximum likelihood estimates being equal to their true values the estimates of the residual series are uncorrelated with zero mean and constant variance
.
You have the option of setting an argument (
exact to Nag_FALSE) so that
nag_tsa_varma_estimate (g13ddc) calculates conditional maximum likelihood estimates (conditional on
). This may be useful if the exact maximum likelihood estimates are close to the boundary of the invertibility region.
You also have the option (see
Section 5) of requesting
nag_tsa_varma_estimate (g13ddc) to constrain elements of the
and
matrices and
vector to have pre-specified values.
4
References
Ansley C F and Kohn R (1986) A note on reparameterising a vector autoregressive moving average model to enforce stationarity J. Statist. Comput. Simulation 24 99–106
Gill P E and Murray W (1972) Quasi-Newton methods for unconstrained optimization J. Inst. Math. Appl. 9 91–108
Shea B L (1987) Estimation of multivariate time series J. Time Ser. Anal. 8 95–110
5
Arguments
- 1:
– IntegerInput
-
On entry: , the number of observed time series.
Constraint:
.
- 2:
– IntegerInput
-
On entry: , the number of observations in each time series.
- 3:
– IntegerInput
-
On entry: , the number of AR parameter matrices.
Constraint:
.
- 4:
– IntegerInput
-
On entry: , the number of MA parameter matrices.
Constraint:
.
is not permitted.
- 5:
– Nag_IncludeMeanInput
-
On entry: , if components of have been estimated and , if all elements of are to be taken as zero.
Constraint:
or .
- 6:
– doubleInput/Output
-
On entry: initial parameter estimates read in row by row in the order
,
.
Thus,
- if ,
must be set equal to an initial estimate of the th element of , for , and ;
- if , must be set equal to an initial estimate of the th element of , and ;
- if , should be set equal to an initial estimate of the th component of (). (If you set to then nag_tsa_varma_estimate (g13ddc) will calculate the mean of the th series and use this as an initial estimate of .)
The first
elements of
par must satisfy the stationarity condition and the next
elements of
par must satisfy the invertibility condition.
If in doubt set all elements of
par to
.
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then all the elements of
par will be overwritten by the latest estimates of the corresponding ARMA parameters.
- 7:
– IntegerInput
-
On entry:
npar is the number of initial parameter estimates.
Constraints:
- if , npar must be set equal to ;
- if , npar must be set equal to .
The total number of observations must exceed the total number of parameters in the model ().
- 8:
– doubleInput/Output
-
On entry:
must be set equal to an initial estimate of the
th element of
. The lower triangle only is needed.
qq must be positive definite. It is strongly recommended that on entry the elements of
qq are of the same order of magnitude as at the solution point. If you set
, for
and
, then
nag_tsa_varma_estimate (g13ddc) will calculate the covariance matrix between the
time series and use this as an initial estimate of
.
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
will contain the latest estimate of the
th element of
. The lower triangle only is returned.
- 9:
– IntegerInput
-
On entry: stride seperating row elements in
qq,
w and
v.
Constraint:
.
- 10:
– const doubleInput
-
On entry: must be set equal to the th component of , for and .
- 11:
– const Nag_BooleanInput
-
On entry:
must be set to Nag_TRUE if
is to be held constant at its input value and Nag_FALSE if
is a free parameter, for
.
If in doubt try setting all elements of
parhld to Nag_FALSE.
- 12:
– Nag_BooleanInput
-
On entry: must be set equal to Nag_TRUE if you wish
nag_tsa_varma_estimate (g13ddc) to compute exact maximum likelihood estimates.
exact must be set equal to Nag_FALSE if only conditional likelihood estimates are required.
- 13:
– IntegerInput
-
On entry: the frequency with which the automatic monitoring function is to be called.
- The ML search procedure is monitored once every iprint iterations and just before exit from the search function.
- The search function is monitored once at the final point.
- The search function is not monitored at all.
- 14:
– doubleInput
-
On entry: the accuracy to which the solution in
par and
qq is required.
If
cgetol is set to
and on exit
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV or
NE_HESS_NOT_POS_DEF, then all the elements in
par and
qq should be accurate to approximately
decimal places. For most practical purposes the value
should suffice. You should be wary of setting
cgetol too small since the convergence criteria may then have become too strict for the machine to handle.
If
cgetol has been set to a value which is less than the
machine precision,
, then
nag_tsa_varma_estimate (g13ddc) will use the value
instead.
- 15:
– IntegerInput
-
On entry: the maximum number of likelihood evaluations to be permitted by the search procedure.
Suggested value:
.
Constraint:
.
- 16:
– IntegerInput
-
On entry: specifies which of the following two quantities are to be printed.
(i) |
table of maximum likelihood estimates and their standard errors (as returned in the output arrays par, qq and cm); |
(ii) |
table of residual series (as returned in the output array v). |
- None of the above are printed.
- (i) only is printed.
- (i) and (ii) are printed.
Constraint:
.
- 17:
– const char *Input
-
On entry: the name of a file to which diagnostic output will be directed. If
outfile is
NULL the diagnostic output will be directed to standard output.
- 18:
– Integer *Output
-
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
niter contains the number of iterations performed by the search function.
- 19:
– double *Output
-
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
rlogl contains the value of the log-likelihood function corresponding to the final point held in
par and
qq.
- 20:
– doubleOutput
-
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
will contain an estimate of the
th component of
, for
and
, corresponding to the final point held in
par and
qq.
- 21:
– doubleOutput
-
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
will contain the estimated first derivative of the log-likelihood function with respect to the
th element in the array
par. If the gradient cannot be computed then all the elements of
g are returned as zero.
- 22:
– doubleOutput
-
On exit: if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK,
NE_G13D_MAXCAL or
NE_HESS_NOT_POS_DEF then
will contain an estimate of the correlation coefficient between the
th and
th elements in the
par array for
,
. If
, then
will contain the estimated standard error of
. If the
th component of
par has been held constant, i.e.,
was set to Nag_TRUE, then the
th row and column of
cm will be set to zero. If the second derivative matrix cannot be computed then all the elements of
cm are returned as zero.
- 23:
– IntegerInput
-
On entry: stride seperating row elements in
cm.
Constraint:
.
- 24:
– NagError *Input/Output
-
The NAG error argument (see
Section 3.7 in How to Use the NAG Library and its Documentation).
6
Error Indicators and Warnings
- NE_ALLOC_FAIL
-
Dynamic memory allocation failed.
See
Section 2.3.1.2 in How to Use the NAG Library and its Documentation for further information.
- NE_BAD_PARAM
-
On entry, argument had an illegal value.
- NE_G13D_AR
-
The initial AR parameter estimates are outside the stationarity region.
To proceed, you must try a different starting point.
- NE_G13D_ARMA
-
On entry, and .
- NE_G13D_BOUND
-
The ML solution is so close to the boundary of either the stationarity region or the invertibility region that
nag_tsa_varma_estimate (g13ddc) cannot evaluate the Hessian matrix. The elements of
cm are set to zero, as are the elements of
g. All other output quantities are correct.
- NE_G13D_DERIV
-
An estimate of the second derivative matrix and the gradient vector at the solution point was computed. Either the Hessian matrix was found to be too ill-conditioned to be evaluated accurately or the gradient vector could not be computed to an acceptable degree of accuracy. The elements of
cm are set to zero, as are the elements of
g. All other output quantities are correct.
- NE_G13D_GRAD
-
The function cannot compute a sufficiently accurate estimate of the gradient vector at the user-supplied starting point. This usually occurs if either the initial parameter estimates are very close to the ML parameter estimates, or you have supplied a very poor estimate of , or the starting point is very close to the boundary of the stationarity or invertibility region. To proceed, you must try a different starting point.
- NE_G13D_MA
-
The initial MA parameter estimates are outside the invertibility region. To proceed, you must try a different starting point.
- NE_G13D_MAX_LOGLIK
-
The conditions for a solution have not all been met, but a point at which the log-likelihood took a larger value could not be found.
Provided that the estimated first derivatives are sufficiently small, and that the estimated condition number of the second derivative (Hessian) matrix, as printed when , is not too large, this error exit may simply mean that, although it has not been possible to satisfy the specified requirements, the algorithm has in fact found the solution as far as the accuracy of the machine permits.
Such a condition can arise, for instance, if cgetol has been set so small that rounding error in evaluating the likelihood function makes attainment of the convergence conditions impossible. If the estimated condition number at the final point is large, it could be that the final point is a solution but that the smallest eigenvalue of the Hessian matrix is so close to zero at the solution that it is not possible to recognize it as a solution. Output quantities were computed at the final point held in par and qq, except that if g or cm could not be computed, in which case they are set to zero.
- NE_G13D_MAXCAL
-
There have been
maxcal log-likelihood evaluations made in the function.
If steady increases in the log-likelihood function were monitored up to the point where this exit occurred, then the exit probably simply occurred because maxcal was set too small, so the calculations should be restarted from the final point held in par and qq. This type of exit may also indicate that there is no maximum to the likelihood surface. Output quantities were computed at the final point held in par and qq, except that if g or cm could not be computed, in which case they are set to zero.
- NE_G13D_START
-
The starting point is too close to the boundary of the admissibility region.
- NE_HESS_NOT_POS_DEF
-
The second-derivative matrix at the solution point is not positive definite. The elements of
cm are set to zero. All other output quantities are correct.
- NE_INT
-
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
- NE_INT_2
-
On entry, and .
Constraint: .
On entry, and .
Constraint: .
- NE_INT_3
-
On entry, , and .
Constraint: .
- 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.
See
Section 2.7.6 in How to Use the NAG Library and its Documentation for further information.
- NE_NO_LICENCE
-
Your licence key may have expired or may not have been installed correctly.
See
Section 2.7.5 in How to Use the NAG Library and its Documentation for further information.
- NE_NOT_CLOSE_FILE
-
Cannot close file .
- NE_NOT_POS_DEF
-
The initial estimate of is not positive definite. To proceed, you must try a different starting point.
- NE_NOT_WRITE_FILE
-
Cannot open file for writing.
7
Accuracy
On exit from
nag_tsa_varma_estimate (g13ddc), if
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV or
NE_HESS_NOT_POS_DEF and
cgetol has been set to
, then all the parameters should be accurate to approximately
decimal places. If
cgetol was set equal to a value less than the
machine precision,
, then all the parameters should be accurate to approximately
.
If
NE_G13D_MAXCAL on exit (i.e.,
maxcal likelihood evaluations have been made but the convergence conditions of the search function have not been satisfied), then the elements in
par and
qq may still be good approximations to the ML estimates. Inspection of the elements of
g may help you determine whether this is likely.
8
Parallelism and Performance
nag_tsa_varma_estimate (g13ddc) is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
nag_tsa_varma_estimate (g13ddc) 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 function. Please also consult the
Users' Note for your implementation for any additional implementation-specific information.
Let and . Local workspace arrays of fixed lengths are allocated internally by nag_tsa_varma_estimate (g13ddc). The total size of these arrays amounts to Integer elements and double elements.
The number of iterations required depends upon the number of parameters in the model and the distance of the user-supplied starting point from the solution.
If the solution lies on the boundary of the admissibility region (stationarity and invertibility region) then
nag_tsa_varma_estimate (g13ddc) may get into difficulty and exit with
NE_G13D_MAX_LOGLIK. If this exit occurs you are advised to either try a different starting point or a different setting for
exact. If this still continues to occur then you are urged to try fitting a more parsimonious model.
You are advised to try and avoid fitting models with an excessive number of parameters since over-parameterisation can cause the maximization problem to become ill-conditioned.
The standardized estimates of the residual series
(denoted by
) can easily be calculated by forming the Cholesky decomposition of
, e.g.,
and setting
.
nag_dpotrf (f07fdc) may be used to calculate the array
g. The components of
which are now uncorrelated at
all lags can sometimes be more easily interpreted.
If your time series model provides a good fit to the data then the residual series should be approximately white noise, i.e., exhibit no serial cross-correlation. An examination of the residual cross-correlation matrices should confirm whether this is likely to be so. You are advised to call
nag_tsa_varma_diagnostic (g13dsc) to provide information for diagnostic checking.
nag_tsa_varma_diagnostic (g13dsc) returns the residual cross-correlation matrices along with their asymptotic standard errors.
nag_tsa_varma_diagnostic (g13dsc) also computes a portmanteau statistic and its asymptotic significance level for testing model adequacy. If
NE_NOERROR or
NE_G13D_BOUND,
NE_G13D_DERIV,
NE_G13D_MAX_LOGLIK or
NE_HESS_NOT_POS_DEF
on exit from
nag_tsa_varma_estimate (g13ddc) then the quantities output
k,
n,
v,
kmax,
ip,
iq,
par,
parhld, and
qq will be suitable for input to
nag_tsa_varma_diagnostic (g13dsc).
10
Example
This example shows how to fit a bivariate AR(1) model to two series each of length . will be estimated and will be constrained to be zero.
10.1
Program Text
Program Text (g13ddce.c)
10.2
Program Data
Program Data (g13ddce.d)
10.3
Program Results
Program Results (g13ddce.r)