I have something I'm attempting to model. A generalized form would be:
yi
= f ( yi-6, yi-7, yi-8, yi-9,yi-10,
yi-11, yi-12, yi-13, yi-14, yi-15,
yi-16, yi-17, xi-6, xi-7, xi-8,
xi-9,xi-10, xi-11, xi-12, xi-13,
xi-14, xi-15, xi-16, xi-17, zi-6,
zi-7, zi-8, zi-9,zi-10, zi-11,
zi-12, zi-13, zyi-14, zi-15, zi-16,
zi-17 )
Proc forecast and proc arima handle the y terms, and don't do too bad. I tried a stepwise linear regression manually lagging the variables.
Is there a proc just for this type of model?