A collection of bandwidth estimators for Kernel Density Estimation
AmiseBW
(data, niterations=25, eps=0.001)[source]¶Bases: conkit.misc.bandwidth.BandwidthBase
Asymptotic Mean Integrated Squared Error (AMISE)
This particular choice of bandwidth recovers all the important features whilst maintaining smoothness. It is a direct implementation of the method used by [1].
[1] | Sadowski, M.I. (2013). Prediction of protein domain boundaries from inverse covariances. |
Attributes
bandwidth |
|
bw |
Methods
curvature (p, x, w) |
|
extended_range (mn, mx, bw[, ext]) |
|
optimal_bandwidth_equation (p, default_bw) |
|
stiffness_integral (p, default_bw[, eps]) |
bandwidth
¶BandwidthBase
[source]¶Bases: abc.ABC
Abstract class for bandwidth calculations
Attributes
bandwidth |
|
bw |
bandwidth
¶bw
¶BowmanBW
(data)[source]¶Bases: conkit.misc.bandwidth.BandwidthBase
Bowman & Azzalini [2] bandwidth calculation
[2] | Bowman, A.W. & Azzalini, A. (1997). Applied Smoothing Techniques for Data Analysis. |
Attributes
bandwidth |
|
bw |
bandwidth
¶LinearBW
(data, threshold=15)[source]¶Bases: conkit.misc.bandwidth.BandwidthBase
Linear [3] implementation
[3] | Sadowski, M.I. (2013). Prediction of protein domain boundaries from inverse covariances. |
Attributes
bandwidth |
|
bw |
bandwidth
¶ScottBW
(data)[source]¶Bases: conkit.misc.bandwidth.BandwidthBase
Scott’s [4] implementation
[4] | Scott, D.W. (1992). Multivariate Density Estimation: Theory, Practice, and Visualization. |
Attributes
bandwidth |
|
bw |
bandwidth
¶