Metadata-Version: 2.4
Name: ValidMLInference
Version: 1.0.7
Summary: This package implements bias correction methods for models estimated using synthetic data
Author-email: Konrad Kurczynski <konrad.kurczynski@yale.edu>, Timothy Christensen <timothy.christensen@yale.edu>
License: MIT
Project-URL: Homepage, https://github.com/KonradKurczynski/ValidMLInference
Classifier: Programming Language :: Python :: 3
Classifier: Operating System :: OS Independent
Requires-Python: >=3.8
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy
Requires-Dist: scipy
Requires-Dist: jax
Requires-Dist: jaxopt
Requires-Dist: numdifftools
Requires-Dist: patsy
Dynamic: license-file

# ValidMLInference

ValidMLInference is a Python package for estimating linear models which use synthetically generated regressors. The bias-correction methods are described in [Battaglia, Christensen, Hansen & Sacher (2024)](https://arxiv.org/abs/2402.15585). 

## Requirements and installation

`ValidMLInference` runs on Python 3.8 and requires a couple of standard numerical packages: numpy, scipy, jax, jaxopt, and numdifftools.You can install `ValidMLInference` by typing ``` > pip install ValidMLInference ```  into the terminal. 

## Using ValidMLInference

To get started with using the package, we recommend looking at the following examples and resources: 
    - [remote_work.ipynb](https://github.com/KonradKurczynski/ValidMLInference/blob/main/remote_work.ipynb) this notebook contains an example of estimating the association between working from home and salaries in job postings using real-world data
    - [synthetic_example.ipynb](https://github.com/KonradKurczynski/ValidMLInference/blob/main/synthetic_example.ipynb) this notebook contains an example showing the performance of the different estimators on synthetic data
    - [functionality.md](https://github.com/KonradKurczynski/ValidMLInference/blob/main/functionality.md) this file contains descriptions of the functions, optional arguments, etc. 
