• Mozart W. Talakua Jurusan Matematika FMIPA Universitas Pattimura
Keywords: optimal bandwidth, Maximum Likelihood Cross-Validation, Unbiased Cross-Validation, Biased Cross-Validation


The choice of the bandwidth h is the main problem of kernel density function. In some situations it might be quite useful to have a set of retes corresponding to different bandwidth.It is necessary to agree which bandwidth is an appropriate one. Cross-Validation (C-V) is a well-known method to optimize the smoothing parameter h. In this research, we will analysis about three methods of Cross-Validation: Maximum Likelihood Cross-Validation, Least-Square Cross-Validation (Unbiased Cross-Validation) and Biased Cross-Validation with bandwidth choice optimal.


Download data is not yet available.


Apostol, T.M.(1967), Calculus Volume I, Wiley International Edition.
Devroye and Penrod,C.S, (1984), The Consistency of Kernel density Estimates, Journal The Annalisis Statistics,Vol 12.,No..4, 1231-1249
Everitt,Brian S. (1994), A Handbook of Statistical analyses using S-Plus, New York.Chapman&Hall
Gaser.T,Muller.H.G. Mamitcsch Society, Kernel for Nonparametric Curve Estimation, Journal of The Royal Statistic Society, Seri B, 47 236-252
Hardle,W (1990), Smoothing Tecnique With Implementation in S., Spinger Verlag, New York.
Hill, P.D. (1985), Kernel Estimation of Distribution Finction, Comm, Statist.Theory Methods, 14, 605-620
Jhon Rice (1984), Bandwidth Choice For Non Parametric Regression, Journal The Anall. Of Statistics, Volume 12, N0.4, 1215-1230
Jeffrey D. Hart and Thomas E. Wehrly, (1992), Kernel regression When The Boundary region in Large, With an Application to testing the Adequacy of Polinomial Medels, Journal of The American Statistical Assosiation, Vol.87.No.420. 1018-1024
Lethold,L (1976), The Calculus With Analitic Geometry, New York Harper International Edition.
How to Cite
M. Talakua, “PENAKSIRAN FUNGSI DENSITAS TIPE KERNEL DENGAN METODE CROSS-VALIDATION (C-V)”, BAREKENG: J. Il. Mat. & Ter., vol. 1, no. 1, pp. 27-31, Mar. 2007.

Most read articles by the same author(s)

1 2 3 > >>