This article demonstrates the applicability of the algorithm of grid search
to the determination of Mark-Houwink (MH) parameters. In the existing meth
ods, the determination of MH parameters involves the use of linear or nonli
near regression. In the cases of using the algorithm of grid search to solv
e the linear regression problem for the determination of MH parameters, the
values of the MH parameters obtained by the grid search method are the sam
e as those obtained by the linear regression method, while in the cases whe
n some nonlinear regression methods have to be used in the existing methods
, using the algorithm of grid search can determine the parameters much more
easily and accurately than using any nonlinear regression methods. In addi
tion, the grid search method is very easy to understand and execute, and th
e accuracy of the results can be exhibited by the value of the least-square
function. (C) 2000 John Wiley & Sons, Inc.