Simulation Study for Boundary Effect Density Estimation
Penulis/Author
KARTIKO (1); Prof. Drs. Suryo Guritno, M.Stats., Ph.D. (2); Prof. Dr.rer.nat. Dedi Rosadi, S.Si.,, M.Sc. (3); Prof. Dr. Abdurakhman, S.Si., M.Si. (4)
Tanggal/Date
1 2013
Kata Kunci/Keyword
Abstrak/Abstract
Density estimation based on data is often use in engineering, finance,
biomedical and social science data. People use parametric fit which
specify the form of the density in advance, which very often is incorrect.
Along with the development of computing software people can easily
do the nonparametric methods, that rely heavily on computing but free
from model assumptions.
Kernel smoothing refers to a general class of techniques for non-
parametric estimation of density functions. For certain univariate set
of data that one wants to display graphically, using kernel function as
the weight, this method can be used and is known as kernel density
estimation.
Estimation in the boundary points suffer a large bias, however a
special treatment is needed. Simulation study is conducted to see that
champernowne transformation can do the job properly.
