dame_flame.utils.post_processing.ATT
The average treatment effect estimate on the treated units in the data
ATT(matching_object)
Uses the matches created by the FLAME and DAME algorithms to provide ATT of the dataset.
Read more about Average Treatment Effect on treated units (ATT) in the User Guide
| Parameter Name | Type | Description |
|---|---|---|
| matching_object | {dame_flame.matching.DAME, dame_flame.matching.FLAME} | The matching object used to run DAME and FLAME. This must be after the .fit() and .predict() methods have been called to create the matches. |
| Return Name | Type | Description |
|---|---|---|
| ATT | {float, np.nan} | A float representing the ATT of the dataset. If no units were matched, then the output will be np.nan. |
Quick Example
import pandas as pd
import dame_flame
df = pd.DataFrame([[0,1,1,1,0,5], [0,1,1,0,0,6], [1,0,1,1,1,7], [1,1,1,1,1,7]],
columns=["x1", "x2", "x3", "x4", "treated", "outcome"])
model = dame_flame.matching.DAME()
model.fit(df)
result = model.predict(df)
att = dame_flame.utils.post_processing.ATT(model)