New keynesian phillips curve in non-linear form - pls help

Hi all,

I need to write a model in its original, non-linear form as I need to run 2nd order approximations. I’m stuck at the stage where firms set prices a la’ calvo, whereby you get to the problem of an infinite sum, as documented in the Dynare User Guide in page 44. However this and other papers all continue with the algebra only after log-linearising the equation.

How can I proceed with the algebra in its non-linear form? Any help or references would be greatly appreciated as I’m running on a tight deadline.

Thanks

Hi, Schmitt-Grohe and Uribe in one of their papers (columbia.edu/~mu2166/chile/chile_appendix.pdf) show how to write this infinite sum recursively.

Many thanks, much appreciated

W