Demo for the DoWhy causal API

We show a simple example of adding a causal extension to any dataframe.

import dowhy.datasets
import dowhy.api
from dowhy.graph import build_graph_from_str

import numpy as np
import pandas as pd

from statsmodels.api import OLS

data = dowhy.datasets.linear_dataset(beta=5,
        num_common_causes=1,
        num_instruments = 0,
        num_samples=1000,
        treatment_is_binary=True)
df = data['df']
df['y'] = df['y'] + np.random.normal(size=len(df)) # Adding noise to data. Without noise, the variance in Y|X, Z is zero, and mcmc fails.
nx_graph = build_graph_from_str(data["dot_graph"])

treatment= data["treatment_name"][0]
outcome = data["outcome_name"][0]
common_cause = data["common_causes_names"][0]
df

	W0	v0	y
0	-0.269430	False	0.362230
1	-0.082544	False	-0.767772
2	-1.814006	False	-0.707189
3	-1.449714	False	-1.707053
4	0.996700	True	2.920611
...	...	...	...
995	-1.426200	False	-1.152158
996	0.799952	True	5.970946
997	0.336987	True	7.368024
998	-1.835364	False	-0.493349
999	0.007103	True	4.539431

1000 rows × 3 columns

# data['df'] is just a regular pandas.DataFrame
df.causal.do(x=treatment,
             variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'},
             outcome=outcome,
             common_causes=[common_cause],
            ).groupby(treatment).mean().plot(y=outcome, kind='bar')

<Axes: xlabel='v0'>

df.causal.do(x={treatment: 1},
              variable_types={treatment:'b', outcome: 'c', common_cause: 'c'},
              outcome=outcome,
              method='weighting',
              common_causes=[common_cause]
              ).groupby(treatment).mean().plot(y=outcome, kind='bar')

<Axes: xlabel='v0'>

cdf_1 = df.causal.do(x={treatment: 1},
              variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'},
              outcome=outcome,
              graph=nx_graph
              )

cdf_0 = df.causal.do(x={treatment: 0},
              variable_types={treatment: 'b', outcome: 'c', common_cause: 'c'},
              outcome=outcome,
              graph=nx_graph
              )

cdf_0

	W0	v0	y	propensity_score	weight
0	0.678214	False	0.638137	0.377125	2.651638
1	0.790310	False	-0.069697	0.358504	2.789370
2	0.641499	False	-0.379419	0.383307	2.608872
3	0.031219	False	-0.845857	0.490134	2.040257
4	0.090265	False	-0.439085	0.479597	2.085085
...	...	...	...	...	...
995	-0.001491	False	-0.046452	0.495976	2.016226
996	-0.504713	False	-0.992955	0.585034	1.709302
997	-0.092837	False	-1.266626	0.512291	1.952015
998	-1.253501	False	-0.125889	0.706513	1.415403
999	-1.231512	False	0.477721	0.703244	1.421981

1000 rows × 5 columns

cdf_1

	W0	v0	y	propensity_score	weight
0	0.112066	True	4.564782	0.524290	1.907342
1	-1.671744	True	4.628525	0.235528	4.245780
2	0.679911	True	6.201627	0.623159	1.604726
3	-1.633031	True	5.480536	0.240545	4.157226
4	0.003191	True	7.543041	0.504860	1.980747
...	...	...	...	...	...
995	0.186091	True	3.327907	0.537462	1.860597
996	0.533387	True	5.117317	0.598275	1.671472
997	-0.211352	True	3.773451	0.466586	2.143230
998	-0.150743	True	5.005458	0.477378	2.094776
999	-1.568709	True	6.005801	0.249041	4.015404

1000 rows × 5 columns

Comparing the estimate to Linear Regression

First, estimating the effect using the causal data frame, and the 95% confidence interval.

(cdf_1['y'] - cdf_0['y']).mean()

$\displaystyle 5.22492144239524$

1.96*(cdf_1['y'] - cdf_0['y']).std() / np.sqrt(len(df))

$\displaystyle 0.091787074923842$

Comparing to the estimate from OLS.

model = OLS(np.asarray(df[outcome]), np.asarray(df[[common_cause, treatment]], dtype=np.float64))
result = model.fit()
result.summary()

OLS Regression Results

Dep. Variable:	y	R-squared (uncentered):	0.920
Model:	OLS	Adj. R-squared (uncentered):	0.920
Method:	Least Squares	F-statistic:	5714.
Date:	Sat, 15 Nov 2025	Prob (F-statistic):	0.00
Time:	06:22:50	Log-Likelihood:	-1429.1
No. Observations:	1000	AIC:	2862.
Df Residuals:	998	BIC:	2872.
Df Model:	2
Covariance Type:	nonrobust

	coef	std err	t	P>|t|	[0.025	0.975]
x1	0.1671	0.031	5.380	0.000	0.106	0.228
x2	5.0148	0.047	106.327	0.000	4.922	5.107

Omnibus:	0.243	Durbin-Watson:	1.959
Prob(Omnibus):	0.885	Jarque-Bera (JB):	0.320
Skew:	-0.024	Prob(JB):	0.852
Kurtosis:	2.927	Cond. No.	1.52

Notes:
[1] R² is computed without centering (uncentered) since the model does not contain a constant.
[2] Standard Errors assume that the covariance matrix of the errors is correctly specified.