Metadata-Version: 2.1
Name: ptfa
Version: 0.1.6
Summary: Probabilistic Targeted Factor Analysis
Author-email: "Miguel C. Herculano" <miguel.herculano@glasgow.ac.uk>, Santiago Montoya-Blandón <Santiago.Montoya-Blandon@glasgow.ac.uk>
Maintainer-email: Santiago Montoya-Blandón <Santiago.Montoya-Blandon@glasgow.ac.uk>, "Miguel C. Herculano" <miguel.herculano@glasgow.ac.uk>
Project-URL: Homepage, https://github.com/smonto2/PTFA
Project-URL: Bug tracking, https://github.com/smonto2/PTFA/issues
Keywords: Partial Least Squares,high-dimensional data,Expectation-Maximization algorithm,missing data,time-series
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE.md
Requires-Dist: numpy
Requires-Dist: scikit-learn

The `ptfa` package introduces routines to obtain factors and loadings from features variables used to predict target variables. This is accomplished through a probabilistic version of Partial Least Squares (PLS) that performs efficient targeted factor extraction. The package provides an array of expectation maximization (EM) algorithms for learning the parameters of this model under a wide range of real-world economic data situations. This includes standard cross-sectional data, and includes extension that can additionally account for missing data (both at-random and in mixed-frequency settings), stochastic volatility or dynamic relationships.
