Bnk_func {QTOCen}

Generate biquadratic kernel weights for a univariate variable

Description

This is the biquadratic kernel function, that weights observations by their distances to the target observation.

Usage

Bnk_func(x0k, Xk, bw.bnk)

Arguments

x0k

Numeric scalar. One univariate covariate value of interest from one observation.

Xk

Numerical vector. The vector of the same covariate from observations

bw.bnk

The bandwith scalar parameter.

Value

This function returns a list of kernel weights with the same length of input Xk.

Note

This function is widely used for generating kernel weights for nonparametrically estimating conditional survival functions. See Section 2.3 of (Wang and Wang 2009).

References

Wang HJ, Wang L (2009). “Locally weighted censored quantile regression.” Journal of the American Statistical Association, 104(487), 1117–1128.

Examples

 
Bnk_func(x0k=0, Xk=c(-5:5), bw.bnk=10)