Compute the Kiefer-Wolfowitz nonparametric maximum likelihood estimator
Project description
Kiefer-Wolfowitz Nonparametric Empirical Bayes
Compute the Kiefer-Wolfowitz nonparametric maximum likelihood estimator for mixtures.
In contrast to the previous approaches, the optimization problem is reformulated into a convex problem by Koenker and Mizera (2014)'s method and efficiently solved by interior-point method.
Making Predictions With No Features - A Basic Usage
Given a training set T = {y_i}, the algorithm provides a way to construct a predictor of future y-values such that the sum of squared errors between observations and predictors is minimized.
Getting Started
Prerequisites
You will need:
- python (>= 3.6)
- pip (>= 19.0.3)
- MOSEK (>=8.1.30)
Important about MOSEK:
- MOSEK is a commercial optimization software. Please visit MOSEK for license information.
- PIP:
pip install -f https://download.mosek.com/stable/wheel/index.html Mosek --user
For different ways of installation, please visit their installation page.
- MOSEK needs to be installed in the GLOBAL environment.
Installing
pip install kwnpeb
Examples
- simple - The basic usage
- bayesball - In-season prediction of batting averages with the 2005 Major League baseball
Contributors
License
This project is licensed under the MIT License - see the LICENSE.md file for details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
| Filename, size | File type | Python version | Upload date | Hashes |
|---|---|---|---|---|
| Filename, size kwnpeb-0.1.11-py3-none-any.whl (4.5 kB) | File type Wheel | Python version py3 | Upload date | Hashes View |
| Filename, size kwnpeb-0.1.11.tar.gz (3.4 kB) | File type Source | Python version None | Upload date | Hashes View |