gbker2d1
Kernel density estimate for bivariate data
Usage
gbker2d [options]
Overview
2D kernel density estimation on an equispaced grid. Data are read from standard input as couple (x,y). The kernel bandwidth if not provided with the option -H is set automatically. Different bandwidths can be specified for x and y.
Options
-
-n -
number of points where the density is computed (default 10) use comma (,) to specify different values for x and y components
-
-H -
set the kernel bandwidth explicitly; use comma (,) to specify different values for x and y components
-
-S -
scale the kernel bandwidth with respect to euristic 'optimal' value; use comma (,) to specify different values for x and y components
-
-K -
choose the kernel to use: 0: Epanenchnikov 1: Rectangular 2: Silverman type 1 3: Silverman type 2 (default 0)
-
-D -
switch off data de-variation procedure
-
-v -
verbose mode
-
-F -
specify the input fields separators (default " \t")
-
-h -
this help
Examples
- gbker2d -n 10,20 -K 1 < file
-
compute the kernel on an equispaced grid 10x20 using a rectangular kernel
Releases
5 releases since Debian 9 (Stretch); the text changed 2 times.