German Credit Logistic Distribution

class pints.toy.GermanCreditLogPDF(x=None, y=None, download=False)

Toy distribution based on a logistic regression model, which takes the form,

\[f(x, y|\beta) \propto \text{exp}(-\sum_{i=1}^{N} \text{log}(1 + \text{exp}(-y_i x_i.\beta)) - \beta.\beta/2\sigma^2)\]

The data \((x, y)\) are a matrix of individual predictors (with 1s in the first column) and responses (1 if the individual should receive credit and -1 if not) respectively; \(\beta\) is a 25x1 vector of coefficients and \(\sigma^2=100\). The dataset here is from [1] but the test problem is defined in [2].

Extends pints.LogPDF.

Parameters:

beta (float) – vector of coefficients of length 25.

References

[1]

“UCI machine learning repository”, 2010. A. Frank and A. Asuncion.

[2]

“The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo”, 2014, M.D. Hoffman and A. Gelman.

data()

Returns data used to fit model.

distance(samples)

Calculates a measure of distance from samples to some characteristic of the underlying distribution.

evaluateS1(beta)

See LogPDF.evaluateS1().

n_parameters()

Returns the dimension of the space this LogPDF is defined over.

sample(n_samples)

Generates independent samples from the underlying distribution.

suggested_bounds()

See ToyLogPDF.suggested_bounds().