Choosing the best value of shape parameter in radial basis functions by Leave-P-Out Cross Validation

Document Type : Research Paper


Faculty of Mathematical Sciences, University of Guilan, Rasht, Iran.


The radial basis functions (RBFs) meshless method has high accuracy for the interpolation of scattered data in high dimensions. Most of the RBFs depend on a parameter, called shape parameter which plays a significant role to specify the accuracy of the RBF method. In this paper, we present three algorithms to choose the optimal value of the shape parameter. These are based on Rippa’s theory to remove data points from the data set and results obtained by Fasshauer and Zhang for the iterative approximate moving least square (AMLS) method. Numerical experiments confirm stable solutions with high accuracy compared to other methods.


  • [1] M. Abbaszadeh and M. Dehghan, Direct meshless local Petrov–Galerkin (DMLPG) method for time-fractional fourth-order reaction–diffusion problem on complex domains, Computers & Mathematics with Applications, 79(3) (2019), 876–888.
  • [2] H. R. Azarboni, M. Keyanpour, and M. Yaghouti, Leave-Two-Out Cross Validation to optimal shape parameter in radial basis functions, Engineering Analysis with Boundary Elements, 100 (2019), 204–210.
  • [3] M. D. Buhmann, Radial basis functions: theory and implementations, Cambridge university press, 2003.
  • [4] R. E. Carlson and T. A. Foley, The parameter R2 in multiquadric interpolation, Computers & Mathematics with Applications, 21(9) (1991), 29–42.
  • [5] W. Chen, Z.-J. Fu, and C.-S. Chen, Recent advances in radial basis function collocation methods, Springer, 2014.
  • [6] M. Dehghan and A. Shokri, A meshless method for numerical solution of the one-dimensional wave equation with an integral condition using radial basis functions, Numerical Algorithms, 52(3) (2009), 461–477.
  • [7] G. E. Fasshauer, Approximate moving least-squares approximation with compactly supported radial weights, Mesh- free methods for partial differential equations, Springer, 105–116, 2003.
  • [8] G. E. Fasshauer, Approximate moving least-squares approximation: A fast and accurate multivariate approxima- tion method, Curve and surface fitting: Saint-Malo, (2002), 139–148.
  • [9] G. E. Fasshauer, Toward approximate moving least squares approximation with irregularly spaced centers, Com- puter Methods in Applied Mechanics and Engineering, 193(12-14) (2004), 1231–1243.
  • [10] G. E. Fasshauer, Meshfree Approximation Methods with MATLAB, World Scientific, 2007.
  • [11] G. E. Fasshauer and J. G. Zhang, Recent results for moving least squares approximation, Geometric Modeling and Computing, (2004), 163–176.
  • [12] G. E. Fasshauer and J. G. Zhang, Scattered data approximation of noisy data via iterated moving least squares, Proceedings of Curve and Surface Fitting: Avignon, (2006), 150–159.
  • [13] G. E. Fasshauer and J. G. Zhang, On choosing “optimal” shape parameters for RBF approximation, Numerical Algorithms, 45(1-4) (2007), 345–368.
  • [14] G. E. Fasshauer and J. G. Zhang, Iterated approximate moving least squares approximation, Advances in Meshfree Techniques, Springer, 221–239, 2007.
  • [15] B. Fornberg and N. Flyer, Solving PDEs with radial basis functions, Acta Numerica, 24 (2015), 215–258.
  • [16] R. Franke, Scattered data interpolation: tests of some methods, Mathematics of Computation, 38(157) (1982), 181–200.
  • [17] M. Ghorbani, Diffuse element kansa method, Applied Mathematical Sciences, 4(12) (2010), 583–594.
  • [18] E. Giannaros, A. Kotzakolios, V. Kostopoulos, and G. Campoli, Hypervelocity impact response of CFRP laminates using smoothed particle hydrodynamics method: Implementation and validation, International Journal of Impact Engineering, 123 (2019), 56–69.
  • [19] R. L. Hardy, Multiquadric equations of topography and other irregular surfaces, Journal of Geophysical Research, 76(8) (1971), 1905–1915.
  • [20] V. R. Hosseini, E. Shivanian, and W. Chen, Local radial point interpolation (MLRPI) method for solving time fractional diffusion-wave equation with damping, Journal of Computational Physics, 312 (2016), 307–332.
  • [21] A. J. Khattak and S. Tirmizi, Application of meshfree collocation method to a class of nonlinear partial differential equations, Engineering analysis with boundary elements, 33(5) (2009), 661–667.
  • [22] P. Lancaster and K. Salkauskas, Surfaces generated by moving least squares methods, Mathematics of computation, 37(155) (1981), 141–158.
  • [23] F. Lanzara, V. Maz’ya, and G. Schmidt, Approximate approximations from scattered data, Journal of Approxi- mation Theory, 145(2) (2007), 141–170.
  • [24] X. Li and H. Dong, Analysis of the element-free Galerkin method for Signorini problems, Applied Mathematics and Computation, 346 (2019), 41–56.
  • [25] Q. Liu, F. Liu, Y. T. Gu, P. Zhuang, J. Chen, and I. Turner, A meshless method based on Point Interpolation Method (PIM) for the space fractional diffusion equation, Applied Mathematics and Computation, 256 (2015), 930–938.
  • [26] V. Maz’ya and G. Schmidt, On quasi-interpolation with non-uniformly distributed centers on domains and man- ifolds, Journal of Approximation Theory, 110(2) (2001), 125–145.
  • [27] V. Maz’ya, A new approximation method and its applications to the calculation of volume potentials. Boundary point method, DFG-Kolloquium des DFG-Forschungsschwerpunktes “Randelementmethoden, 1991.
  • [28] C. A. Micchelli, Interpolation of scattered data: distance matrices and conditionally positive definite functions, Constructive Approximation, 2 (1986), 11–22.
  • [29]  S. Rippa, An algorithm for selecting a good value for the parameter c in radial basis function interpolation, Advances in Computational Mathematics, 11(2-3) (1999), 193–210.
  • [30] M. Scheuerer, An alternative procedure for selecting a good value for the parameter c in RBF-interpolation, Advances in Computational Mathematics, 34(1) (2011), 105–126.
  • [31] I. J. Schoenberg, Metric spaces and completely monotone functions, Annals of Mathematics, 39 (1938), 811–841.