ARMAX Approach

[joshua.crsb]

Hello,

First - I apologize for the naivete of the question - I am just learning the program and know enough to know when to ask for help.

I have been working on this for about a week now, and can't seem to get it right. For an assignment I have been asked to forecast of Sales, (datafiles are given for both Sales and CPI) using the ARMAX approach. However, I can't seem to get a probable equation. Currently the equation I have is, "sales/(cpi/100) c ar(1) ma(1)", and the results aren't likely. I was wondering if I am structuring the equation correctly for ARMAX - if not what will fix it?

Thanks,
JC
:D

[joshua.crsb]

Quick revision to the equation:

I have estimated a variety of additional equations. However, many of the equations I run I get the following prompt in the stats table the formula generates, "Estimated AR process is nonstationary". A couple of the equations that this pops up in are: "sales/(cpi/100) c ar(1) ma(1) ma(2) ma(3)",
"sales/(cpi/100) c ar(1) ar(2) ma(1) ma(2)", and "sales/(cpi/100) c ar(1) ar(2) ar(3) ma(1) ma(2)" ---- I have been going to AR(3) and MA(3)...I don't know if this is a good idea.

Thanks again!!!

[startz]

You might want to begin by looking at a graph of your data and seeing if it has a trend. You might also want to test if the data is nonstationary.