On Some Computational Aspects of Real Business Cycle Theory