ker: Type of kernel function

In bbemkr: Bayesian bandwidth estimation for multivariate kernel regression with Gaussian error

Description

For data that have infinite support, Gaussian kernel is suggested. For data that have [-1, 1] support, other types of kernel can be used.

Usage

ker(u, kerntype = c("Gaussian", "Epanechnikov", "Quartic", 
                    "Triweight", "Triangular", "Uniform"))

Arguments

u	A numeric object
kerntype	Type of kernel function

Details

Oftentimes, we deal with numeric values of infinite support, Gaussian kernel is commonly used. However, Epanechnikov kernel is the optimal kernel as measured by Mean Integrated Square Error. The difference among kernel functions is minor, but the influence of bandwidths is vital.

Value

Kernel value

Author(s)

Han Lin Shang

References

J. Fan and I. Gijbels (1996) Local Polynomial Modelling and Its Application. Chapman and Hall, London.

Q. Li and J. Racine (2007) Nonparametric Econometrics: Theory and Practice. Princeton University Press, New Jersey.

See Also

np_gibbs, gibbs_admkr_nw, gibbs_admkr_erro

bbemkr documentation built on May 1, 2019, 10:53 p.m.