The RATS Software Forum

Dear Tom and Friends on the Forum,
I have been trying to generate annual growth rate as Kim, Sei-Wan, A. V. Mollick and Kiseok Nam in their paper " Common nonlinearities in long-horizon stock returns: Evidence from the G-7 stock markets". Global Finance Journal: 19 (2008):19-31. As they claim that the they calculte the annual return is different with Sarantis (2001). Details shown on page 22 of this paper in the section 2 of annuallized returns from overlaps of monthly prices. I try to do it using RATS but I still doubt about my codes. Is anyone can help me to do this in a efficient way?

Thanks.

I think you're misinterpreting what they said. The annual growth rates are the same as in Sarantis---it's the choice of the "lag" that's different. The annual growth rates are

set growth = log(p/p{12})

Sarantis uses growth{1} as the lag in his analysis, while the authors of this paper use growth{12}. You can either use growth{12} directly, or define

set glagged = growth{12}

and then use glagged.