User Tools

Site Tools


perseus:user:activities:matrixprocessing:basic:densityestimationprocessing

Density estimation

General

Brief description

The density of data points in two dimensions is calculated. Each data point is smoothed out by a suitable Gaussian kernel.

Output: A copy of the input matrix with two numerical columns added containing the density information.

Parameters

x

Selected expression columns for the first dimension of the generated density map(s) (default: first expression column in the matrix). Multiple columns can be chosen for each dimension leading to the creation of multiple density maps, but the number of columns have to be the same in both dimensions.

y

Selected expression columns for the second dimension of the generated density map(s) (default: second expression column in the matrix). Multiple columns can be chosen for each dimension leading to the creation of multiple density maps, but the number of columns have to be the same in both dimensions.

Number of points

The specified number of points defines the resolution of the density map (default: 300). It reflects the number of pixel per dimension.

Distribution type

Each data point is smoothed out by a suitable Gaussian kernel, which is defined in the “Distribution type” (default: P(x,y). The function can be selected out of a predefined list:

  • P(x,y)
  • P(y|x)
  • P(x|y)
  • P(x,y)/(P(x)*P(y))


Parameter window

Perseus pop-up window: Basic -> Density estimation

perseus/user/activities/matrixprocessing/basic/densityestimationprocessing.txt · Last modified: 2015/11/23 14:47 (external edit)