azcausal 0.2.0

Installation

To install the current release, please execute:

pip install git+https://github.com/amazon-science/azcausal.git

Usage

import numpy as np

from azcausal.core.error import Placebo
from azcausal.core.panel import Panel
from azcausal.core.parallelize import Pool
from azcausal.data import CaliforniaProp99
from azcausal.estimators.panel.sdid import SDID
from azcausal.util import to_matrices

# load an example data set with the columns Year, State, PacksPerCapita, treated.
df = CaliforniaProp99().load()

# convert to matrices where the index represents each Year (time) and each column a state (unit)
data = to_matrices(df, "Year", "State", "PacksPerCapita", "treated")

# create a panel object to access observations conveniently
panel = Panel(outcome="PacksPerCapita", intervention="treated", data=data)

# initialize an estimator object, here synthetic difference in difference (sdid)
estimator = SDID()

# run the estimator
result = estimator.fit(panel)

# create a process pool for parallelization
pool = Pool(mode="thread", progress=True)

# run the error validation method
method = Placebo(n_samples=11)
estimator.error(result, method, parallelize=pool)

# print out information about the estimate
print(result.summary(title="CaliforniaProp99"))

Estimators

• Difference-in-Difference (DID): Simple implementation of the well-known Difference-in-Difference estimator.

• Synthetic Difference-in-Difference (SDID): Arkhangelsky, Dmitry Athey, Susan Hirshberg, David A. Imbens, Guido W. Wager, Stefan Synthetic Difference-in-Differences American Economic Review 111 12 4088-4118 2021 10.1257/aer.20190159 https://www.aeaweb.org/articles?id=10.1257/aer.20190159. Implementation based on https://synth-inference.github.io/synthdid/

Contact

Feel free to contact me if you have any questions: