Time Series Analysis Package
ARMAORDERS Method
rc = TSA.ARMAORDERS (y, dif, <'method'>, <p>, <q>, <perror>, <porders>, <qorders>);
The ARMAORDERS function performs tests to tentatively identify the autoregressive and moving average orders of mixed autoregressive moving average models.
Required Arguments
You must specify the following argument:
- y
specifies the times series array to test.
- dif
specifies either an array of positive integers or a positive integer that is used for differencing. The default value is 0.
Optional Arguments
You can also specify the following arguments, separated by commas. If you want to use a default value for any of these arguments, enter a space for it.
- 'method'
-
specifies the method of tentative order selection.
You can specify the following values within single quotation marks:
- ESACF
specifies the extended sample autocorrelation function.
- MINIC
specifies the minimum information criterion.
- SCAN
specifies the squared canonical correlations.
The default value is MINIC.
- p
-
specifies the autoregressive order range, where p is an array of two nonnegative integers that define the minimum and maximum values.
By default, p is the array [0,5].
- q
-
specifies the moving average order range, where q is an array of two nonnegative integers that define the minimum and maximum values.
By default, q is the array [0,5].
- perror
-
specifies the autoregressive orders used to estimate the error series for the MINIC method, where perror is an array of two nonnegative integers that define the minimum and maximum values.
By default, perror is the array [
(p), (p)+(q)].
Returned Values
The ARMAORDERS function returns the following values:
- rc
-
returns one of the following scalar return codes:
rc Termination Reason 0 Success < 0 Computational failure
Optional Returned Values
You can also specify the following arguments, separated by commas, to request additional returned values. If you do not want the value to be returned, enter a space for it.
- porders
returns the recommended autoregressive orders.
- qorders
returns the recommended moving average orders.
Example
This example uses the TSMODEL procedure to tentatively identify the autoregressive and moving average orders for the time series Air:
proc tsmodel data=mylib.air outscalar=mylib.outscalars;
id date interval=month;
var air;
outscalars porders qorders;
require tsa;
submit;
declare object TSA(tsa);
porders = 0;
qorders = 0;
Array P[2]/nosymbols; P[1]=0; P[2]=5;
Array Q[2]/nosymbols; Q[1]=0; Q[2]=5;
rc=TSA.ARMAORDERS(air, 0, 'SCAN', P, Q, , porders, qorders);
rc=TSA.ARMAORDERS(air, 1, 'ESACF', P, Q, , porders, qorders);
rc=TSA.ARMAORDERS(air, 1, 'MINIC', P, Q, ,porders, qorders);
endsubmit;
run;