SPSS help and econometrics?

Question

Hi I'm really stuck with SPSS.
My hypothesis states that an increase to independent variable (oil prices) will lead to a decrease in dependent variable (Real GDp). Should I be running a linear regression model (OLS) to see if this is the case?
I also want to add adjusted time-lags to see if oil has a higher correlation to GDP when there is a year delay. Does anyone know how I would do this through SPSS?

Answer 1

Hi,

The best way to do it is to use Granger causality test, which is not actually supported by SPSS, you can use EViews instead. Or, you can try to search for SPSS syntax (macros) for Granger test.

Answer 2

You can run a linear regression via Analyze> Regression > Linear or, in most recent versions, you could use Regression > Automatic Linear Modeling. Lagged variables can be computed using COMPUTE and the LAG function.



But you to consider the reverse relationship: perhaps changes in Real GDP affect the price of oil by reducing demand. That calls for a simultaneous equations model. In SPSS, you can estimate these with Two-Stage Least Squares, but you need to understand something about this class of models first.

Answer 3

Granger Causality Spss