09-07-2011 03:59 PM
Hello, I'm not sure if this is the right place to be requesting help with a problem for school but I figured I'd try.
Anyway, I am trying to use SAS and familiar with how to use if for multiple and logistic regression, however, I am struggling using it now for Time Series.
I am supposed to create a sample of length 200 from the time series model and then I am supposed to:
1. Plot the Series
2. Give the corresponding ACF and PACF plots.
3. Identify it as ARMA (p,q) for appropriate values of p and q.
The model (with subscripts listed as lowercase letters) the sample is supposed to run off is:
Yt= 1.5Y(t-1) - 0.56Y(t-2) + Et
here is the code I entered into SAS, I recieve the error message LIBNAME TS is not assigned.
do N=1 to 300;
e = normal (0822016280);
y= 1.5*y_1-0.56*y_2 + e;
time = -100 + n;
if n > 100 then output;
keep time y;
Someone please help me. Thanks a lot in advance.
09-07-2011 04:04 PM
In SAS, a dot in a data set name is interpretted as a two-level name: library.member.
Assuming you want to create the data in the WORK library, just use DATA TSARMA11 without the dot (or use an underscore, if you want).
09-07-2011 04:13 PM
Hey Rick, thanks for your reply. I removed the period so my first line now reads as you wrote. I now get the error message:
NOTE: The data set WORK.TSARMA11 has 200 observations and 2 variables.
NOTE: DATA statement used (Total process time):
real time 0.04 seconds
cpu time 0.00 seconds