12-07-2016 08:26 AM - edited 12-07-2016 08:31 AM
Can someone explain to me. what this code is doing?
basically, i want to display the year 2008 and 2010 with the different month sold in the products. My code do not recognise xlog. Please help.
proc arima data=temp_orders;
estimate q=(1)(12) noint method=ml;
forecast id=orderdate interval=month printall out=b;
I got this error message :ERROR: Forecasting was not performed because estimation was not done.
12-07-2016 08:33 AM
There is no xlog option for the identify statement in proc arima, see the documentation: http://support.sas.com/documentation/cdl/en/etsug/68148/HTML/default/viewer.htm#etsug_arima_syntax04...
I suggest you consult the documentation of a certain piece of SAS before applying it.
12-07-2016 08:37 AM
References are always useful. It looks like your code came from this example in the PROC ARIMA documentation.
In the example, XLOG is the response variable in the data set. So you need to replace XLOG by the name of the variable in your data that contains the time series.
The (1, 12) specifies the lags. In this case, the model has lags of 1 and 12.
12-07-2016 10:18 PM
Yes. Rick is right. xlog is log(numunits) . Did you see the code before proc arima ? there is xlog=log(numunits); Log it is to handle big number and make data more suitable for ARIMA .