Learn R Programming

kedd (version 1.0.2)

h.mcv: Modified Cross-Validation for Bandwidth Selection

Description

The (S3) generic function h.mcv computes the modified cross-validation bandwidth selector of r'th derivative of kernel density estimator one-dimensional.

Usage

h.mcv(x, ...)
## S3 method for class 'default':
h.mcv(x, deriv.order = 0, lower = 0.1 * hos, upper = 2 * hos, 
         tol = 0.1 * lower, kernel = c("gaussian", "epanechnikov", "triweight", 
         "tricube", "biweight", "cosine"), ...)

Arguments

x
vector of data values.
deriv.order
derivative order (scalar).
lower, upper
range over which to minimize. The default is almost always satisfactory. hos (Over-smoothing) is calculated internally from an kernel, see details.
tol
the convergence tolerance for optimize.
kernel
a character string giving the smoothing kernel to be used, with default "gaussian".
...
further arguments for (non-default) methods.

Value

  • xdata points - same as input.
  • data.namethe deparsed name of the x argument.
  • nthe sample size after elimination of missing values.
  • kernelname of kernel to use
  • deriv.orderthe derivative order to use.
  • hvalue of bandwidth parameter.
  • min.mcvthe minimal MCV value.

Details

h.mcv modified cross-validation implements for choosing the bandwidth $h$ of a r'th derivative kernel density estimator. Stute (1992) proposed a so-called modified cross-validation (MCV) in kernel density estimator. This method can be extended to the estimation of derivative of a density, the essential idea based on approximated the problematic term by the aid of the Hajek projection (see Stute 1992). The minimization criterion is defined by: $$MCV(h;r) = \frac{R\left(K^{(r)}\right)}{nh^{2r+1}} + \frac{(-1)^{r}}{n(n-1)h^{2r+1}}\sum_{i=1}^{n} \sum_{j=1;j \neq i}^{n} \varphi^{(r)} \left(\frac{X_{j}-X_{i}}{h}\right)$$ whit $$\varphi^{(r)}(c) = \left(K^{(r)} \ast K^{(r)} - K^{(2r)} - \frac{\mu_{2}(K)}{2}K^{(2r+2)} \right)(c)$$ and $K^{(r)} \ast K^{(r)} (x)$ is the convolution of the r'th derivative kernel function $K^{(r)}(x)$ (see kernel.conv and kernel.fun); $R\left(K^{(r)}\right) = \int_{R} K^{(r)}(x)^{2} dx$ and $\mu_{2}(K) = \int_{R}x^{2} K(x) dx$. The range over which to minimize is hos Oversmoothing bandwidth, the default is almost always satisfactory. See George and Scott (1985), George (1990), Scott (1992, pp 165), Wand and Jones (1995, pp 61).

References

Heidenreich, N. B., Schindler, A. and Sperlich, S. (2013). Bandwidth selection for kernel density estimation: a review of fully automatic selectors. Advances in Statistical Analysis. Stute, W. (1992). Modified cross validation in density estimation. Journal of Statistical Planning and Inference, 30, 293--305.

See Also

plot.h.mcv.

Examples

Run this code
## Derivative order = 0

h.mcv(kurtotic,deriv.order = 0)

## Derivative order = 1

h.mcv(kurtotic,deriv.order = 1)

Run the code above in your browser using DataLab