Which proc to use

Mark_WGU — Thu, 10 Oct 2013 17:40:23 GMT

I have something I'm attempting to model. A generalized form would be:

y_i
= f ( y_i-6, y_i-7, y_i-8, y_i-9,y_i-10,
y_i-11, y_i-12, y_i-13, y_i-14, y_i-15,
y_i-16, y_i-17, x_i-6, x_i-7, x_i-8,
x_i-9,x_i-10, x_i-11, x_i-12, x_i-13,
x_i-14, x_i-15, x_i-16, x_i-17, z_i-6,
z_i-7, z_i-8, z_i-9,z_i-10, z_i-11,
z_i-12, z_i-13, zy_i-14, z_i-15, z_i-16,
z_i-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?

Re: Which proc to use

Mit — Thu, 10 Oct 2013 23:45:12 GMT

What about proc glm ?

topic Re: Which proc to use in Statistical Procedures

Which proc to use

Re: Which proc to use