NAG Library Function Document
nag_durbin_watson_stat (g02fcc)
1 Purpose
nag_durbin_watson_stat (g02fcc) calculates the Durbin–Watson statistic, for a set of residuals, and the upper and lower bounds for its significance.
2 Specification
#include <nag.h> |
#include <nagg02.h> |
void |
nag_durbin_watson_stat (Integer n,
Integer p,
const double res[],
double *d,
double *pdl,
double *pdu,
NagError *fail) |
|
3 Description
For the general linear regression model
where |
is a vector of length of the dependent variable,
is a by matrix of the independent variables,
is a vector of length of unknown arguments, |
and |
is a vector of length of unknown random errors. |
The residuals are given by
and the fitted values,
, can be written as
for a
by
matrix
. Note that when a mean term is included in the model the sum of the residuals is zero. If the observations have been taken serially, that is
can be considered as a time series, the Durbin–Watson test can be used to test for serial correlation in the
, see
Durbin and Watson (1950),
Durbin and Watson (1951) and
Durbin and Watson (1971).
The Durbin–Watson statistic is
Positive serial correlation in the
will lead to a small value of
while for independent errors
will be close to
. Durbin and Watson show that the exact distribution of
depends on the eigenvalues of the matrix
where the matrix
is such that
can be written as
and the eigenvalues of the matrix
are
, for
.
However bounds on the distribution can be obtained, the lower bound being
and the upper bound being
where the
are independent standard Normal variables. The lower tail probabilities associated with these bounds,
and
, are computed by
nag_prob_durbin_watson (g01epc). The interpretation of the bounds is that, for a test of size (significance)
, if
the test is significant, if
the test is not significant, while if
and
no conclusion can be reached.
The above probabilities are for the usual test of positive auto-correlation. If the alternative of negative auto-correlation is required, then a call to
nag_prob_durbin_watson (g01epc) should be made with the argument
d taking the value of
; see
Newbold (1988).
4 References
Durbin J and Watson G S (1950) Testing for serial correlation in least squares regression. I Biometrika 37 409–428
Durbin J and Watson G S (1951) Testing for serial correlation in least squares regression. II Biometrika 38 159–178
Durbin J and Watson G S (1971) Testing for serial correlation in least squares regression. III Biometrika 58 1–19
Granger C W J and Newbold P (1986) Forecasting Economic Time Series (2nd Edition) Academic Press
Newbold P (1988) Statistics for Business and Economics Prentice–Hall
5 Arguments
- 1:
n – IntegerInput
On entry: , the number of residuals.
Constraint:
.
- 2:
p – IntegerInput
On entry: , the number of independent variables in the regression model, including the mean.
Constraint:
.
- 3:
res[n] – const doubleInput
On entry: the residuals, .
Constraint:
the mean of the residuals , where .
- 4:
d – double *Output
On exit: the Durbin–Watson statistic, .
- 5:
pdl – double *Output
On exit: lower bound for the significance of the Durbin–Watson statistic, .
- 6:
pdu – double *Output
On exit: upper bound for the significance of the Durbin–Watson statistic, .
- 7:
fail – NagError *Input/Output
-
The NAG error argument (see
Section 3.6 in the Essential Introduction).
6 Error Indicators and Warnings
- NE_ALLOC_FAIL
-
Dynamic memory allocation failed.
- NE_BAD_PARAM
-
On entry, argument had an illegal value.
- NE_INT
-
On entry, .
Constraint: .
- NE_INT_2
-
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.
- NE_RESID_IDEN
-
On entry, all residuals are identical.
- NE_RESID_MEAN
-
On entry, the mean of
res is not approximately
,
.
7 Accuracy
The probabilities are computed to an accuracy of at least decimal places.
8 Parallelism and Performance
Not applicable.
If the exact probabilities are required, then the first
eigenvalues of
can be computed and
nag_prob_lin_chi_sq (g01jdc) used to compute the required probabilities with the argument
c set to
and the argument
d set to the Durbin–Watson statistic
.
10 Example
A set of residuals are read in and the Durbin–Watson statistic along with the probability bounds are computed and printed.
10.1 Program Text
Program Text (g02fcce.c)
10.2 Program Data
Program Data (g02fcce.d)
10.3 Program Results
Program Results (g02fcce.r)