Empirical Likelihood emplike

Introduction

Empirical likelihood is a method of nonparametric inference and estimation that lifts the obligation of having to specify a family of underlying distributions. Moreover, empirical likelihood methods do not require re-sampling but still uniquely determine confidence regions whose shape mirrors the shape of the data. In essence, empirical likelihood attempts to combine the benefits of parametric and nonparametric methods while limiting their shortcomings. The main difficulties of empirical likelihood is the computationally intensive methods required to conduct inference. statsmodels.emplike attempts to provide a user-friendly interface that allows the end user to effectively conduct empirical likelihood analysis without having to concern themselves with the computational burdens.

Currently, emplike provides methods to conduct hypothesis tests and form confidence intervals for descriptive statistics. Empirical likelihood estimation and inference in a regression, accelerated failure time and instrumental variable model are currently under development.

References

The main reference for empirical likelihood is:

Owen, A.B. "Empirical Likelihood." Chapman and Hall, 2001.

Examples

In [1]: import numpy as np

In [2]: import statsmodels.api as sm

# Generate Data
In [3]: x = np.random.standard_normal(50)

# initiate EL
In [4]: el = sm.emplike.DescStat(x)

# confidence interval for the mean
In [5]: el.ci_mean()
Out[5]: (-0.32649017605906, 0.2670353125042696)

# test variance is 1
In [6]: el.test_var(1)
Out[6]: (0.21983144508529795, 0.639168378769406)

Module Reference

descriptive.DescStat(endog)

Returns an instance to conduct inference on descriptive statistics via empirical likelihood.

descriptive.DescStatUV(endog)

A class to compute confidence intervals and hypothesis tests involving mean, variance, kurtosis and skewness of a univariate random variable.

descriptive.DescStatMV(endog)

A class for conducting inference on multivariate means and correlation.