Big Difference

[alim]

Dear Researcher

I am very new in this discussion site. I need help. I have two variables i.e., 1. Australian hard coal production 2. Australian steam coal export. When I check their correlation at level. The R square is .90. It means highly correlated. Both data are stationary in first difference and they are not cointegrated at level. According to the theory, if the variables are not cointegrated, we can only run the regression with the stationary varibles (e.g., first difference variable). But when I am estimating the stationary variables. The R square is showing 0.01. Between two R squares result is big difference (.90 and .01) Which is totall unbelievable. In Regression, at level in addition one tonne hard coal production increases, the Australian steam coal export increases by .403 and Regression in stationary variables shows that in addition one tonne Australian hard coal production increases, the Australian steam coal export increases by .03 If any body knows the reason or where am I doing wrong, please inform me via my following e-mail address. I appreciate your help.

Thanking you

With regard

Md. Liakat Ali
E-mail: mdliakat@yahoo.com