latentcor 0.2.5

Fast Computation of Latent Correlations for Mixed Data

• Free software: GNU General Public License v3

• Documentation: https://latentcor-py.readthedocs.io.

Introduction

latentcor is an Python package for estimation of latent correlations with mixed data types (continuous, binary, truncated, and ternary) under the latent Gaussian copula model. For references on the estimation framework, see

• Fan, J., Liu, H., Ning, Y., and Zou, H. (2017), “High Dimensional Semiparametric Latent Graphical Model for Mixed Data.” JRSS B. Continuous/binary types.

• Quan X., Booth J.G. and Wells M.T. “Rank-based approach for estimating correlations in mixed ordinal data.” arXiv. Ternary type.

• Yoon G., Carroll R.J. and Gaynanova I. (2020). “Sparse semiparametric canonical correlation analysis for data of mixed types.” Biometrika. Truncated type for zero-inflated data.

• Yoon G., Müller C.L. and Gaynanova I. (2021). “Fast computation of latent correlations.”. Approximation method of computation, see math framework for details.

Statement of need

No Python software package is currently available that allows accurate and fast correlation estimation from mixed variable data in a unifying manner. The Python package latentcor, introduced here, thus represents the first stand-alone Python package for computation of latent correlation that takes into account all variable types (continuous/binary/ordinal/zero-inflated), comes with an optimized memory footprint, and is computationally efficient, essentially making latent correlation estimation almost as fast as rank-based correlation estimation.

Installation

The easiest way to install latentcor is using pip.

pip install latentcor

Example

Let’s import gen_data, get_tps and latentcor from latentcor.

from latentcor import gen_data, get_tps, latentcor

First, we will generate a pair of variables with different types using a sample size n=100 which will serve as example data. Here first variable will be ternary, and second variable will be continuous.

simdata = gen_data(n = 100, tps = ["ter", "con"])
print(simdata['X'][ : 6, : ])

Then we can estimate the latent correlation matrix based on these 2 variables using latentcor function.

estimate = latentcor(simdata['X'], tps = ["ter", "con"])
print(estimate['R'])

Community Guidelines

• Contributions and suggestions to the software are always welcome. Please consult our contribution guidelines prior to submitting a pull request.

• Report issues or problems with the software using github’s issue tracker.

• The easiest way to replicate development environment of latentcor is using pip:

pip install -r requirements_dev.txt

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

History

0.1.0 (2021-12-28)

• First version.

0.1.1 (2022-01-06)

• Fix some typos.

0.1.2 (2022-01-06)

• Fix some bug on use_nearPD argument in function latentcor.

0.1.3 (2022-01-07)

• Fix syntax errors for jupyter-execute in README.txt.

0.1.4 (2022-05-23)

• Fix error for continuous estimation.

0.2.0 (2022-08-16)

• Increase maximum iteration for positive definiteness adjustment.

• Make function outputs as dictionary.

0.2.1 (2022-08-22)

• Make output latent correlation matrix as pandas.DataFrame.

• Polish output heatmap.

0.2.2 (2022-08-22)

• Update README file.

0.2.3 (2022-08-22)

• Correct update history.

0.2.4 (2022-09-07)

• Correct incompatible versions.

0.2.5 (2023-11-05)

• Regenerate interpolants for approximation method and fix version compatibility for Python 3.7.