appelpy: Applied Econometrics Library for Python
About 👁️
appelpy is the Applied Econometrics Library for Python. It seeks to bridge the gap between the software options that have a simple syntax (such as Stata) and other powerful options that use Python's object-oriented programming as part of data modelling workflows.
Econometric modelling and general regression analysis in Python have never been easier!
The library builds upon the functionality of the 'vanilla' Python data stack (e.g. Pandas, Numpy, etc.) and other libraries such as Statsmodels.
Documentation
See the functionality of the library at https://appelpy.readthedocs.io/
Get started with the 10 Minutes to Appelpy tutorial:
: interactive experience of the 10 Minutes to Appelpy tutorial via Binder.
: static render of the 10 Minutes to Appelpy notebook.
🥧 Why it's as easy as pie
Here is a flavour of a basic OLS regression done through appelpy, supposing you have data sitting in a Pandas dataframe df and want to model the dependent variable api00 on three other variables:
from appelpy.linear_model import OLS model1 = OLS(df, ['api00'], ['acs_k3', 'meals', 'full']).fit() model1.results_output # returns summary results
The key information is sitting in the model1 object, but there is much more functionality that can be done with it. These are more things that can be done via one line of code:
- Diagnostics can be called from the object: e.g. produce a P-P plot via
model1.diagnostic_plot('pp_plot') - Model selection statistics: e.g. find the root mean square error of the model from
model1.model_selection_stats - Standardized model estimates:
model1.results_output_standardized
🍏 What inspired it?
The simple syntax of software such as Stata. With the data loaded, a regression model summary can be returned by a one-line command:
regress api00 acs_k3 meals full
However with the simplicity comes a few disadvantages: Stata is not open-source software; the workflows are tricky with modern business problems; lacks the benefits of object-oriented programming.
Statsmodels is a powerful Python library that addresses some of those disadvantages, but with that power comes a considerable learning curve and clunkiness. Here is the code for the same regression:
import statsmodels.api as sm model1 = sm.OLS(df['api00'], sm.add_constant(df['acs_k3', 'meals', 'full'])).fit() results1 = model1.summary() # returns summary results
It can get much more unwieldy than that. The model results object is brilliant as it can be printed in different formats (plaintext, Latex, etc.)... but that is only the starting point. How do I diagnose the regression model itself? How do I get standardized estimates? That's where it becomes more complicated.
appelpy simply aims to achieve a sweet spot between both approaches.
Installation ⏲️
pip install appelpy
Supported for Python 3.6 and higher versions.
Dependencies 🖇️
- pandas>=0.24
- jinja2
- scipy
- numpy>=1.16
- statsmodels>=0.9
- patsy
- seaborn>=0.9
- matplotlib>=3
Licence ⚖️
Modified BSD (3-clause)