lmdiag 0.3.7

Diagnostic Plots for Lineare Regression Models. Similar to plot.lm in R.

Python Library providing Diagnostic Plots for Lineare Regression Models. (Like plot.lm in R.)

I built this, because I missed the diagnostics plots of R for a university project. There are some substitutions in Python for individual charts, but they are spread over different libraries and sometimes don’t show the exact same. My implementation tries to copycat the R-plots, but I didn’t reimplement the R-code: The charts are just based on available documentation.

Installation

Available in PyPi: https://pypi.org/project/lmdiag/

• Using pip: pip install lmdiag

• Using pipenv: pipenv install lmdiag

Usage

The plots need a fitted Linear Regression Model created by statsmodels as input.

Fitted Model from Linearmodels <https://bashtage.github.io/linearmodels/doc/index.html> should also work, however, that’s not tested very well.

Example

(See also the more extensive Example Notebook)

import numpy as np
import matplotlib.pyplot as plt
import statsmodels.api as sm
import lmdiag

%matplotlib inline  # In Jupyter

# Generate sample model
np.random.seed(20)
predictor = np.random.normal(size=30, loc=20, scale=3)
response = 5 + 5 * predictor + np.random.normal(size=30)
X = sm.add_constant(predictor)
lm = sm.OLS(response, X).fit()

# Plot chart matrix (and enlarge figure)
plt.figure(figsize=(10,7))
lmdiag.plot(lm);

Methods

• Draw matrix of all plots:

  lmdiag.plot(lm)

• Draw individual plots:

  lmdiag.resid_fit(lm)

  lmdiag.q_q(lm)

  lmdiag.scale_loc(lm)

  lmdiag.resid_lev(lm)

• Print useful descriptions for interpretations:

  lmdiag.info() (for all plots)

  lmdiag.info('<method name>') (for individual plot)

Development

Disclaimer

This is my very first public python library. Don’t expect everything to work smoothly. I’m happy to receive useful feedback or pull requests.

‘Programming Language :: Python :: 3.6’,

Certification

Packaging and Upload to PyPi

• pipenv run rstcheck README.rst (check syntax)

• rm -rf ./dist (delete old builds)

• python setup.py sdist

• python setup.py bdist_wheel

• twine upload dist/*

• Then new release on github…