Improved Statistical Method for Computing Mean Dimensions of Polymer Molecules

Abstract

Further work on the dimensions of coiling type polymer molecules has been carried out with an electronic digital computer using nonintersecting random walks as molecular models. A modified method has been devised which significantly reduces the serious attrition of samples. The procedure, applied to walks in the two‐choice square lattice, consists of adding strides of four steps instead of adding single steps one by one. The strides are so chosen that a valid statistical sample can be attained while eliminating most of the attrition attributable to closures of four‐membered rings. The method shows promise of being applicable to problems of a more general nature.