Investigation of shape parameter for exponential weight function in moving least squares method

The Moving Least Square (MLS) Method is an approach which is used in meshfree solutions and data approximation. In the formulation of MLS, the exponential weight function is applied to influence the formation of continuous approximated functions. The governing factor in the weight function is the shape parameter, yet the effect of shape parameter to the data approximation is unclear. Therefore the objective of the paper is to investigate the effect of shape parameter to the equation formed via MLS. The three examples are studied and the results show that there is no direct relationship between shape function and prediction accuracy