Likelihood Estimation for Observational data with Python
Project description
Data with uncertain, missing, censored, and correlated values are commonplace in many research fields including astronomy. Unfortunately, such data are often treated in an ad hoc way potentially resulting in inconsistent parameter estimates. Furthermore, in a realistic setting, the variables of interest or their errors may have non-normal distributions which complicates the modeling. LEO-Py uses a novel technique to compute the likelihood function for such data sets. This approach employs Gaussian copulas to decouple the correlation structure of variables and their marginal distributions resulting in a flexible method to compute likelihood functions of data in the presence of measurement uncertainty, censoring, and missing data.
If you use any version of this code, please properly reference the code paper: Feldmann, R. (2019) “LEO-Py: Estimating likelihoods for correlated, censored, and uncertain data with given marginal distributions”, Astronomy & Computing
Copyright 2019 University of Zurich, Robert Feldmann
LEO-Py requires a working python3.5 installation or later to run.
Before installing LEO-Py, you may want to set up a virtual environment python -mvenv /path/to/new/virtual/environment and activate it via source /path/to/new/virtual/environment/bin/activate.
To install LEO-Py from a repository:
Download the source code from <https://github.com/rfeldmann/leopy>.
Go to the package directory.
Run python setup.py install.
To install LEO-Py via PyPI:
Run pip install leopy-stat.
Note, example scripts and the documentation are not installed in this case.
To test the installation:
Run python setup.py test from the package directory (if installed from source).
Go to the ‘site-packages’ directory and run python -m pytest leopy (if installed via PyPI).
To access the code documentation (if installed from source):
Run python setup.py build_html from the package directory.
Open ./build/sphinx/html/index.html to read the documentation.
Using leopy is very simple and consists of 4 steps:
Load the module (import leopy).
Create an observational data set (leopy.Observation).
Create a likelihood instance (leopy.Likelihood).
Call function p() of the likelihood instance.
For instance, a minimal example is:
import pandas as pd from leopy import Observation, Likelihood d = {'v0': [1, 2], 'e_v0': [0.1, 0.2], 'v1': [3, 4], 'e_v1': [0.1, 0.1]} obs = Observation(pd.DataFrame(d), 'testdata') # Reading dataset 'testdata' and extracting 2 variables (['v0', 'v1']) # Errors of different observables are assumed to be uncorrelated l = Likelihood(obs, p_true='lognorm', p_cond='norm') l.p([0.5, 0.7], [1, 2], shape_true=[[1.4], [2.]]) # array([[0.0441545], # [0.0108934]])
Further examples are provided in the ‘paper’ and ‘examples’ sub-directories (if installed from source).
LEO-Py is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
LEO-Py is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with LEO-Py. If not, see <https://www.gnu.org/licenses/>.
Project 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.
Source Distribution
Built Distribution
Hashes for leopy_stat-0.9.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3872a2e76fc6dc1aa0f6b42bba6719aa7cc0232ef6aed6bd9461c17ba48c0010 |
|
MD5 | 66960f06a9083b87f39f417a8a427755 |
|
BLAKE2b-256 | f8da22aa5d1e3b4c856ef23e3544eab386bd40ec0becd4aaf7ff0c17a87fd447 |