Goodness-of-Fit Tests using Kernelized Stein Discrepancy

An adaptation of Kernelized Stein Discrepancy, this package provides a goodness-of-fit test of whether a given i.i.d. sample is drawn from a given distribution. It works for any distribution once its score function (the derivative of log-density) can be provided. This method is based on "A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluation" by Liu, Lee, and Jordan, available at < http://arxiv.org/abs/1602.03253>.


KSD --- This package provides a goodness-of-fit test of whether a given i.i.d. sample {xi} is drawn from a given distribution. It works for any distribution once its score function (the derivative of log-density) ∇xlogp(x) can be provided. This method is based on ``A Kernelized Stein Discrepancy for Goodness-of-fit Tests and Model Evaluation'' by Liu, Lee, and Jordan, available at http://arxiv.org/abs/1602.03253.

The main function of this package is KSD, which estimates Kernelized Stein Discrepancy. Parameters include :

  • x Sample of size Num_Instance x Num_Dimension
  • score_function Score funtion (∇xlogp(x)) : takes x as input and output a column vector of size Num_Instance X Dimension. User may use pryr package to pass in a function that only takes in dataset as parameter, or user may also pass in computed score for a given dataset.
  • kernel Type of kernel (default = 'rbf')
  • width Bandwidth of the kernel
  • nboot Bootstrap sample size

Other methods are also in this package, including various demos and examples.

KSD requires user to provide a score function to be used for computation. For example usage and exploration, a gmm class is provided in the pacakge, which allow test KSD using gaussian mixture model.

Consider the following examples :

  1. We define a gmm, generate random data using the model, and test the null hypothesis that the data comes from the model. Obviously, the result will depend on the model and the amount of added random noise.
# Pass in a dataset generated by Gaussian distribution,
# pass in computed score rather than score function
 
library(KSD)
library(pryr)
#> Warning: package 'pryr' was built under R version 3.2.5
 
model <- gmm()
X <- rgmm(model, n=100)
score_function = scorefunctiongmm(model=model, X=X)
result <- KSD(X,score_function=score_function)
result$p
#> [1] 0.899
  1. We follow similar pattern, but in this example, we use pryr library to define a score function like a function handle in matlab, in which we pass in model as part of the function.
# Pass in a dataset generated by Gaussian distribution,
# use pryr package to pass in score function
library(KSD)
library(pryr)
model <- gmm()
X <- rgmm(model, n=100)
score_function = pryr::partial(scorefunctiongmm, model=model)
result <- KSD(X,score_function=score_function)
result$p
#> [1] 0.899

Premade demos include the following (Note that these demos require additional libraries)

demo_iris()
demo_normal_performance()
demo_simple_gaussian()
demo_simple_gamma()
demo_gmm()
demo_gmm_multi()

A sample run of demo_iris :

library(KSD)
library(datasets)
library(ggplot2)
#> Warning: package 'ggplot2' was built under R version 3.2.3
library(gridExtra)
#> Warning: package 'gridExtra' was built under R version 3.2.3
library(mclust)
#> Warning: package 'mclust' was built under R version 3.2.3
#> Package 'mclust' version 5.1
#> Type 'citation("mclust")' for citing this R package in publications.
library(pryr)
 
demo_iris()
#> [1] "Fitting GMM with 3 clusters"
#> [1] "Average p value : 0.366"

Currently, the code is available at https://github.com/MinHyung-Kang/KSD/ More download options will be available after CRAN submission.

Minhyung(dot)Daniel(dot)Kang(at)gmail(dot)com

News

Initial submission to CRAN

Name of score function parameter in KSD was changed from 'score_q' to 'score_function'. Descriptions have been added.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("KSD")

1.0.0 by Min Hyung Kang, a year ago


Browse source code at https://github.com/cran/KSD


Authors: Min Hyung Kang [aut, cre], Qiang Liu [aut]


Documentation:   PDF Manual  


MIT + file LICENSE license


Imports pryr, graphics, stats

Suggests datasets, ggplot2, gridExtra, mclust, mvtnorm


See at CRAN