Skip to content

Simple Randomization Estimators

This page documents estimators that work with simple randomized experimental designs where treatment assignment is completely randomized.

These estimators leverage pre-treatment covariates through distributional regression frameworks to improve the precision of distributional treatment effect estimates. The key methodological contribution is using machine learning techniques for variance reduction while maintaining validity as long as nuisance components are reasonably well estimated.

Byambadalai et al. (2024)1 propose a regression adjustment method that incorporates covariates into distributional regression, enabling deeper insights beyond average treatment effects by estimating full distributional treatment effects in randomized experiments.

SimpleDistributionEstimator

Bases: SimpleStratifiedDistributionEstimator

A class for computing the empirical distribution function and distributional treatment effects using simple (unadjusted) estimation methods.

This estimator computes Distribution Treatment Effects (DTE), Probability Treatment Effects (PTE), and Quantile Treatment Effects (QTE) without using machine learning models for adjustment. It provides a baseline approach suitable when treatment assignment is random or when covariate adjustment is not needed.

Example: 

import numpy as np
from dte_adj import SimpleDistributionEstimator

# Generate sample data
X = np.random.randn(1000, 5)
D = np.random.binomial(1, 0.5, 1000)  # Random treatment
Y = X[:, 0] + 2 * D + np.random.randn(1000)

# Fit simple estimator
estimator = SimpleDistributionEstimator()
estimator.fit(X, D, Y)

# Compute treatment effects
locations = np.linspace(Y.min(), Y.max(), 20)
dte, lower, upper = estimator.predict_dte(1, 0, locations)
pte, pte_lower, pte_upper = estimator.predict_pte(1, 0, locations)

Initializes the SimpleDistributionEstimator.

Returns: SimpleDistributionEstimator: An instance of the estimator.

Source code in dte_adj/simple.py 
42
43
44
45
46
47
48
def __init__(self):
    """Initializes the SimpleDistributionEstimator.

    Returns:
        SimpleDistributionEstimator: An instance of the estimator.
    """
    super().__init__()

predict_dte 

predict_dte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Distribution Treatment Effects (DTE) based on the estimator for the distribution function.

The DTE measures the difference in cumulative distribution functions between treatment groups at specified locations. It quantifies how treatment affects the probability of observing outcomes below each threshold.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. locations (np.ndarray, optional): Scalar values to be used for computing the cumulative distribution. If None, evenly-spaced locations spanning the observed outcome range are generated automatically. The number of points is determined from data size and distribution via np.histogram_bin_edges(outcomes, bins='auto'). The actual array used is stored on self.last_locations. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. variance_type (str, optional): Variance type to be used to compute confidence intervals. Available values are "moment", "simple", and "uniform". Defaults to "moment". n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected DTEs (np.ndarray): Treatment effect estimates at each location - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleDistributionEstimator

# Generate sample data
X = np.random.randn(1000, 5)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + np.random.randn(1000)

# Fit estimator
estimator = SimpleDistributionEstimator()
estimator.fit(X, D, Y)

# Compute DTE
locations = np.linspace(Y.min(), Y.max(), 20)
dte, lower, upper = estimator.predict_dte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    locations=locations,
    variance_type="moment"
)

print(f"DTE shape: {dte.shape}")  # Should match locations.shape
print(f"Average DTE: {dte.mean():.3f}")

Source code in dte_adj/base.py 
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
def predict_dte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Distribution Treatment Effects (DTE) based on the estimator for the distribution function.

    The DTE measures the difference in cumulative distribution functions between treatment groups
    at specified locations. It quantifies how treatment affects the probability of observing
    outcomes below each threshold.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        locations (np.ndarray, optional): Scalar values to be used for computing the cumulative
            distribution. If None, evenly-spaced locations spanning the observed outcome range
            are generated automatically. The number of points is determined from data size and
            distribution via ``np.histogram_bin_edges(outcomes, bins='auto')``. The actual array
            used is stored on ``self.last_locations``.
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        variance_type (str, optional): Variance type to be used to compute confidence intervals.
            Available values are "moment", "simple", and "uniform". Defaults to "moment".
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected DTEs (np.ndarray): Treatment effect estimates at each location
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleDistributionEstimator

        # Generate sample data
        X = np.random.randn(1000, 5)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + np.random.randn(1000)

        # Fit estimator
        estimator = SimpleDistributionEstimator()
        estimator.fit(X, D, Y)

        # Compute DTE
        locations = np.linspace(Y.min(), Y.max(), 20)
        dte, lower, upper = estimator.predict_dte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            locations=locations,
            variance_type="moment"
        )

        print(f"DTE shape: {dte.shape}")  # Should match locations.shape
        print(f"Average DTE: {dte.mean():.3f}")
        ```
    """
    if locations is None:
        locations = _infer_default_locations(self.outcomes, for_intervals=False)
    self.last_locations = locations
    return self._compute_dtes(
        target_treatment_arm,
        control_treatment_arm,
        locations,
        alpha,
        variance_type,
        n_bootstrap,
        display_progress,
    )

predict_pte 

predict_pte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Probability Treatment Effects (PTE) based on the estimator for the distribution function.

The PTE measures the difference in probability mass between treatment groups for intervals defined by consecutive location pairs. It quantifies how treatment affects the probability of observing outcomes within specific ranges.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. locations (np.ndarray, optional): Scalar values defining interval boundaries for probability computation. For each interval (locations[i], locations[i+1]], the PTE is computed. If None, boundaries spanning the observed outcome range are generated automatically with the left endpoint placed just below outcomes.min() so that minimum-valued samples fall inside the first interval. The number of boundaries is determined from data size and distribution via np.histogram_bin_edges(outcomes, bins='auto'). The actual array used is stored on self.last_locations. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. variance_type (str, optional): Variance type to be used to compute confidence intervals. Available values are "moment", "simple", and "uniform". Defaults to "moment". n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected PTEs (np.ndarray): Treatment effect estimates for each interval, shape (len(locations)-1,) - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleDistributionEstimator

# Generate sample data
X = np.random.randn(1000, 5)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + np.random.randn(1000)

# Fit estimator
estimator = SimpleDistributionEstimator()
estimator.fit(X, D, Y)

# Define interval boundaries
locations = np.array([-2, -1, 0, 1, 2])  # Creates intervals: (-2,-1], (-1,0], (0,1], (1,2]

# Compute PTE
pte, lower, upper = estimator.predict_pte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    locations=locations,
    variance_type="moment"
)

print(f"PTE shape: {pte.shape}")  # Should be (4,) for 4 intervals
print(f"Interval effects: {pte}")

Source code in dte_adj/base.py 
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
def predict_pte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Probability Treatment Effects (PTE) based on the estimator for the distribution function.

    The PTE measures the difference in probability mass between treatment groups for intervals
    defined by consecutive location pairs. It quantifies how treatment affects the probability
    of observing outcomes within specific ranges.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        locations (np.ndarray, optional): Scalar values defining interval boundaries for
            probability computation. For each interval (locations[i], locations[i+1]], the PTE
            is computed. If None, boundaries spanning the observed outcome range are generated
            automatically with the left endpoint placed just below ``outcomes.min()`` so that
            minimum-valued samples fall inside the first interval. The number of boundaries is
            determined from data size and distribution via
            ``np.histogram_bin_edges(outcomes, bins='auto')``. The actual array used is stored
            on ``self.last_locations``.
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        variance_type (str, optional): Variance type to be used to compute confidence intervals.
            Available values are "moment", "simple", and "uniform". Defaults to "moment".
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected PTEs (np.ndarray): Treatment effect estimates for each interval,
              shape (len(locations)-1,)
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleDistributionEstimator

        # Generate sample data
        X = np.random.randn(1000, 5)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + np.random.randn(1000)

        # Fit estimator
        estimator = SimpleDistributionEstimator()
        estimator.fit(X, D, Y)

        # Define interval boundaries
        locations = np.array([-2, -1, 0, 1, 2])  # Creates intervals: (-2,-1], (-1,0], (0,1], (1,2]

        # Compute PTE
        pte, lower, upper = estimator.predict_pte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            locations=locations,
            variance_type="moment"
        )

        print(f"PTE shape: {pte.shape}")  # Should be (4,) for 4 intervals
        print(f"Interval effects: {pte}")
        ```
    """
    if locations is None:
        locations = _infer_default_locations(self.outcomes, for_intervals=True)
    self.last_locations = locations
    return self._compute_ptes(
        target_treatment_arm,
        control_treatment_arm,
        locations,
        alpha,
        variance_type,
        n_bootstrap,
        display_progress,
    )

predict_qte 

predict_qte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    quantiles: Optional[ndarray] = None,
    alpha: float = 0.05,
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Quantile Treatment Effects (QTE) based on the estimator for the distribution function.

The QTE measures the difference in quantiles between treatment groups, providing insights into how treatment affects different parts of the outcome distribution. For stratified estimators, the computation properly accounts for strata.

Variance is estimated by stratified bootstrap: indices are resampled with replacement within each stratum independently, which preserves per-stratum sample sizes and reflects the covariate-adaptive randomization (CAR) design. For estimators without strata (single stratum), this degenerates to a plain bootstrap.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. quantiles (np.ndarray, optional): Quantiles used for QTE. Defaults to [0.1, 0.2, ..., 0.9]. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected QTEs (np.ndarray): Treatment effect estimates at each quantile - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleStratifiedDistributionEstimator

# Generate stratified sample data
X = np.random.randn(1000, 5)
strata = np.random.choice([0, 1, 2], size=1000)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + 0.5 * strata + np.random.randn(1000)

# Fit stratified estimator
estimator = SimpleStratifiedDistributionEstimator()
estimator.fit(X, D, Y, strata)

# Compute QTE at specific quantiles
quantiles = np.array([0.25, 0.5, 0.75])  # 25th, 50th, 75th percentiles
qte, lower, upper = estimator.predict_qte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    quantiles=quantiles,
    n_bootstrap=100
)

print(f"QTE at quantiles {quantiles}: {qte}")
print(f"Median effect (50th percentile): {qte[1]:.3f}")

Source code in dte_adj/base.py 
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
def predict_qte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    quantiles: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Quantile Treatment Effects (QTE) based on the estimator for the distribution function.

    The QTE measures the difference in quantiles between treatment groups, providing insights
    into how treatment affects different parts of the outcome distribution. For stratified
    estimators, the computation properly accounts for strata.

    Variance is estimated by stratified bootstrap: indices are resampled with replacement
    within each stratum independently, which preserves per-stratum sample sizes and reflects
    the covariate-adaptive randomization (CAR) design. For estimators without strata
    (single stratum), this degenerates to a plain bootstrap.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        quantiles (np.ndarray, optional): Quantiles used for QTE. Defaults to [0.1, 0.2, ..., 0.9].
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected QTEs (np.ndarray): Treatment effect estimates at each quantile
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleStratifiedDistributionEstimator

        # Generate stratified sample data
        X = np.random.randn(1000, 5)
        strata = np.random.choice([0, 1, 2], size=1000)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + 0.5 * strata + np.random.randn(1000)

        # Fit stratified estimator
        estimator = SimpleStratifiedDistributionEstimator()
        estimator.fit(X, D, Y, strata)

        # Compute QTE at specific quantiles
        quantiles = np.array([0.25, 0.5, 0.75])  # 25th, 50th, 75th percentiles
        qte, lower, upper = estimator.predict_qte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            quantiles=quantiles,
            n_bootstrap=100
        )

        print(f"QTE at quantiles {quantiles}: {qte}")
        print(f"Median effect (50th percentile): {qte[1]:.3f}")
        ```
    """
    if quantiles is None:
        quantiles = np.arange(1, 10) / 10
    if np.any((quantiles <= 0) | (quantiles >= 1)):
        raise ValueError("quantiles must be in the open interval (0, 1)")

    qte = self._compute_qtes(
        target_treatment_arm,
        control_treatment_arm,
        quantiles,
        self.covariates,
        self.treatment_arms,
        self.outcomes,
        self.strata,
    )

    # Precompute stratum indices for stratified bootstrap.
    # When there is a single stratum this is equivalent to plain bootstrap.
    unique_strata = np.unique(self.strata)
    strata_indices = [np.where(self.strata == s)[0] for s in unique_strata]

    qtes = np.zeros((n_bootstrap, qte.shape[0]))
    bootstrap_iter = range(n_bootstrap)
    if display_progress:
        bootstrap_iter = tqdm(bootstrap_iter, desc="Bootstrap QTE")
    for b in bootstrap_iter:
        bootstrap_indexes = np.concatenate(
            [
                np.random.choice(idx, size=len(idx), replace=True)
                for idx in strata_indices
            ]
        )

        qtes[b] = self._compute_qtes(
            target_treatment_arm,
            control_treatment_arm,
            quantiles,
            self.covariates[bootstrap_indexes],
            self.treatment_arms[bootstrap_indexes],
            self.outcomes[bootstrap_indexes],
            self.strata[bootstrap_indexes],
        )

    qte_var = qtes.var(axis=0)

    qte_lower = qte + norm.ppf(alpha / 2) * np.sqrt(qte_var)
    qte_upper = qte + norm.ppf(1 - alpha / 2) * np.sqrt(qte_var)

    return qte, qte_lower, qte_upper

predict 

predict(
    treatment_arm: int,
    locations: ndarray,
    display_progress: bool = True,
) -> np.ndarray

Compute cumulative distribution values.

Args: treatment_arm (int): The index of the treatment arm. locations (np.ndarray): Scalar values to be used for computing the cumulative distribution. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: np.ndarray: Estimated cumulative distribution values for the input.

Source code in dte_adj/base.py 
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
def predict(
    self, treatment_arm: int, locations: np.ndarray, display_progress: bool = True
) -> np.ndarray:
    """
    Compute cumulative distribution values.

    Args:
        treatment_arm (int): The index of the treatment arm.
        locations (np.ndarray): Scalar values to be used for computing the cumulative distribution.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        np.ndarray: Estimated cumulative distribution values for the input.
    """
    if self.outcomes is None:
        raise ValueError(
            "This estimator has not been trained yet. Please call fit first"
        )

    if treatment_arm not in self.treatment_arms:
        raise ValueError(
            f"This target treatment arm was not included in the training data: {treatment_arm}"
        )

    return self._compute_cumulative_distribution(
        treatment_arm,
        locations,
        self.covariates,
        self.treatment_arms,
        self.outcomes,
        display_progress=display_progress,
    )[0]

fit 

fit(
    covariates: ArrayLike,
    treatment_arms: ArrayLike,
    outcomes: ArrayLike,
) -> SimpleDistributionEstimator

Set parameters.

Args: covariates: Pre-treatment covariates. treatment_arms: The index of the treatment arm. outcomes: Scalar-valued observed outcome.

Returns: SimpleDistributionEstimator: The fitted estimator.

Source code in dte_adj/simple.py 
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
def fit(
    self, covariates: ArrayLike, treatment_arms: ArrayLike, outcomes: ArrayLike
) -> SimpleDistributionEstimator:
    """
    Set parameters.

    Args:
        covariates: Pre-treatment covariates.
        treatment_arms: The index of the treatment arm.
        outcomes: Scalar-valued observed outcome.

    Returns:
        SimpleDistributionEstimator: The fitted estimator.
    """
    covariates = _convert_to_ndarray(covariates)
    treatment_arms = _convert_to_ndarray(treatment_arms)
    outcomes = _convert_to_ndarray(outcomes)

    if covariates.shape[0] != treatment_arms.shape[0]:
        raise ValueError("The shape of covariates and treatment_arm should be same")

    if covariates.shape[0] != outcomes.shape[0]:
        raise ValueError("The shape of covariates and outcome should be same")

    self.covariates = covariates
    self.treatment_arms = treatment_arms
    self.outcomes = outcomes
    self.strata = np.zeros(len(self.covariates))

    return self

AdjustedDistributionEstimator

Bases: AdjustedStratifiedDistributionEstimator

A class for computing distribution treatment effects using machine learning adjustment.

This estimator uses cross-fitting with ML models to adjust for confounding when computing Distribution Treatment Effects (DTE), Probability Treatment Effects (PTE), and Quantile Treatment Effects (QTE). It provides more precise estimates when treatment assignment depends on observed covariates.

Example: 

import numpy as np
from sklearn.ensemble import RandomForestClassifier
from dte_adj import AdjustedDistributionEstimator

# Generate confounded data
X = np.random.randn(1000, 5)
treatment_prob = 1 / (1 + np.exp(-(X[:, 0] + X[:, 1])))
D = np.random.binomial(1, treatment_prob, 1000)
Y = X.sum(axis=1) + 2 * D + np.random.randn(1000)

# Fit adjusted estimator
base_model = RandomForestClassifier(n_estimators=100)
estimator = AdjustedDistributionEstimator(base_model, folds=3)
estimator.fit(X, D, Y)

# Compute adjusted treatment effects
locations = np.linspace(Y.min(), Y.max(), 20)
dte, lower, upper = estimator.predict_dte(1, 0, locations, variance_type="moment")

Initializes the AdjustedDistributionEstimator.

Args: base_model (scikit-learn estimator): The base model implementing used for conditional distribution function estimators. The model should implement fit(data, targets) and predict_proba(data). folds (int): The number of folds for cross-fitting. is_multi_task(bool): Whether to use multi-task learning. If True, your base model needs to support multi-task prediction (n_samples, n_features) -> (n_samples, n_targets).

Returns: AdjustedDistributionEstimator: An instance of the estimator.

Source code in dte_adj/stratified.py 
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
def __init__(self, base_model: Any, folds=3, is_multi_task=False):
    """
    Initializes the AdjustedDistributionEstimator.

    Args:
        base_model (scikit-learn estimator): The base model implementing used for conditional distribution function estimators. The model should implement fit(data, targets) and predict_proba(data).
        folds (int): The number of folds for cross-fitting.
        is_multi_task(bool): Whether to use multi-task learning. If True, your base model needs to support multi-task prediction (n_samples, n_features) -> (n_samples, n_targets).

    Returns:
        AdjustedDistributionEstimator: An instance of the estimator.
    """
    if (not hasattr(base_model, "predict")) and (
        not hasattr(base_model, "predict_proba")
    ):
        raise ValueError(
            "Base model should implement either predict_proba or predict"
        )
    self.base_model = base_model
    self.folds = folds
    self.is_multi_task = is_multi_task
    super().__init__()

predict_dte 

predict_dte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Distribution Treatment Effects (DTE) based on the estimator for the distribution function.

The DTE measures the difference in cumulative distribution functions between treatment groups at specified locations. It quantifies how treatment affects the probability of observing outcomes below each threshold.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. locations (np.ndarray, optional): Scalar values to be used for computing the cumulative distribution. If None, evenly-spaced locations spanning the observed outcome range are generated automatically. The number of points is determined from data size and distribution via np.histogram_bin_edges(outcomes, bins='auto'). The actual array used is stored on self.last_locations. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. variance_type (str, optional): Variance type to be used to compute confidence intervals. Available values are "moment", "simple", and "uniform". Defaults to "moment". n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected DTEs (np.ndarray): Treatment effect estimates at each location - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleDistributionEstimator

# Generate sample data
X = np.random.randn(1000, 5)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + np.random.randn(1000)

# Fit estimator
estimator = SimpleDistributionEstimator()
estimator.fit(X, D, Y)

# Compute DTE
locations = np.linspace(Y.min(), Y.max(), 20)
dte, lower, upper = estimator.predict_dte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    locations=locations,
    variance_type="moment"
)

print(f"DTE shape: {dte.shape}")  # Should match locations.shape
print(f"Average DTE: {dte.mean():.3f}")

Source code in dte_adj/base.py 
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
def predict_dte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Distribution Treatment Effects (DTE) based on the estimator for the distribution function.

    The DTE measures the difference in cumulative distribution functions between treatment groups
    at specified locations. It quantifies how treatment affects the probability of observing
    outcomes below each threshold.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        locations (np.ndarray, optional): Scalar values to be used for computing the cumulative
            distribution. If None, evenly-spaced locations spanning the observed outcome range
            are generated automatically. The number of points is determined from data size and
            distribution via ``np.histogram_bin_edges(outcomes, bins='auto')``. The actual array
            used is stored on ``self.last_locations``.
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        variance_type (str, optional): Variance type to be used to compute confidence intervals.
            Available values are "moment", "simple", and "uniform". Defaults to "moment".
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected DTEs (np.ndarray): Treatment effect estimates at each location
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleDistributionEstimator

        # Generate sample data
        X = np.random.randn(1000, 5)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + np.random.randn(1000)

        # Fit estimator
        estimator = SimpleDistributionEstimator()
        estimator.fit(X, D, Y)

        # Compute DTE
        locations = np.linspace(Y.min(), Y.max(), 20)
        dte, lower, upper = estimator.predict_dte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            locations=locations,
            variance_type="moment"
        )

        print(f"DTE shape: {dte.shape}")  # Should match locations.shape
        print(f"Average DTE: {dte.mean():.3f}")
        ```
    """
    if locations is None:
        locations = _infer_default_locations(self.outcomes, for_intervals=False)
    self.last_locations = locations
    return self._compute_dtes(
        target_treatment_arm,
        control_treatment_arm,
        locations,
        alpha,
        variance_type,
        n_bootstrap,
        display_progress,
    )

predict_pte 

predict_pte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Probability Treatment Effects (PTE) based on the estimator for the distribution function.

The PTE measures the difference in probability mass between treatment groups for intervals defined by consecutive location pairs. It quantifies how treatment affects the probability of observing outcomes within specific ranges.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. locations (np.ndarray, optional): Scalar values defining interval boundaries for probability computation. For each interval (locations[i], locations[i+1]], the PTE is computed. If None, boundaries spanning the observed outcome range are generated automatically with the left endpoint placed just below outcomes.min() so that minimum-valued samples fall inside the first interval. The number of boundaries is determined from data size and distribution via np.histogram_bin_edges(outcomes, bins='auto'). The actual array used is stored on self.last_locations. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. variance_type (str, optional): Variance type to be used to compute confidence intervals. Available values are "moment", "simple", and "uniform". Defaults to "moment". n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected PTEs (np.ndarray): Treatment effect estimates for each interval, shape (len(locations)-1,) - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleDistributionEstimator

# Generate sample data
X = np.random.randn(1000, 5)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + np.random.randn(1000)

# Fit estimator
estimator = SimpleDistributionEstimator()
estimator.fit(X, D, Y)

# Define interval boundaries
locations = np.array([-2, -1, 0, 1, 2])  # Creates intervals: (-2,-1], (-1,0], (0,1], (1,2]

# Compute PTE
pte, lower, upper = estimator.predict_pte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    locations=locations,
    variance_type="moment"
)

print(f"PTE shape: {pte.shape}")  # Should be (4,) for 4 intervals
print(f"Interval effects: {pte}")

Source code in dte_adj/base.py 
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
def predict_pte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    locations: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    variance_type="moment",
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Probability Treatment Effects (PTE) based on the estimator for the distribution function.

    The PTE measures the difference in probability mass between treatment groups for intervals
    defined by consecutive location pairs. It quantifies how treatment affects the probability
    of observing outcomes within specific ranges.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        locations (np.ndarray, optional): Scalar values defining interval boundaries for
            probability computation. For each interval (locations[i], locations[i+1]], the PTE
            is computed. If None, boundaries spanning the observed outcome range are generated
            automatically with the left endpoint placed just below ``outcomes.min()`` so that
            minimum-valued samples fall inside the first interval. The number of boundaries is
            determined from data size and distribution via
            ``np.histogram_bin_edges(outcomes, bins='auto')``. The actual array used is stored
            on ``self.last_locations``.
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        variance_type (str, optional): Variance type to be used to compute confidence intervals.
            Available values are "moment", "simple", and "uniform". Defaults to "moment".
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected PTEs (np.ndarray): Treatment effect estimates for each interval,
              shape (len(locations)-1,)
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleDistributionEstimator

        # Generate sample data
        X = np.random.randn(1000, 5)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + np.random.randn(1000)

        # Fit estimator
        estimator = SimpleDistributionEstimator()
        estimator.fit(X, D, Y)

        # Define interval boundaries
        locations = np.array([-2, -1, 0, 1, 2])  # Creates intervals: (-2,-1], (-1,0], (0,1], (1,2]

        # Compute PTE
        pte, lower, upper = estimator.predict_pte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            locations=locations,
            variance_type="moment"
        )

        print(f"PTE shape: {pte.shape}")  # Should be (4,) for 4 intervals
        print(f"Interval effects: {pte}")
        ```
    """
    if locations is None:
        locations = _infer_default_locations(self.outcomes, for_intervals=True)
    self.last_locations = locations
    return self._compute_ptes(
        target_treatment_arm,
        control_treatment_arm,
        locations,
        alpha,
        variance_type,
        n_bootstrap,
        display_progress,
    )

predict_qte 

predict_qte(
    target_treatment_arm: int,
    control_treatment_arm: int,
    quantiles: Optional[ndarray] = None,
    alpha: float = 0.05,
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]

Compute Quantile Treatment Effects (QTE) based on the estimator for the distribution function.

The QTE measures the difference in quantiles between treatment groups, providing insights into how treatment affects different parts of the outcome distribution. For stratified estimators, the computation properly accounts for strata.

Variance is estimated by stratified bootstrap: indices are resampled with replacement within each stratum independently, which preserves per-stratum sample sizes and reflects the covariate-adaptive randomization (CAR) design. For estimators without strata (single stratum), this degenerates to a plain bootstrap.

Args: target_treatment_arm (int): The index of the treatment arm of the treatment group. control_treatment_arm (int): The index of the treatment arm of the control group. quantiles (np.ndarray, optional): Quantiles used for QTE. Defaults to [0.1, 0.2, ..., 0.9]. alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05. n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing: - Expected QTEs (np.ndarray): Treatment effect estimates at each quantile - Lower bounds (np.ndarray): Lower confidence interval bounds - Upper bounds (np.ndarray): Upper confidence interval bounds

Example: 

import numpy as np
from dte_adj import SimpleStratifiedDistributionEstimator

# Generate stratified sample data
X = np.random.randn(1000, 5)
strata = np.random.choice([0, 1, 2], size=1000)
D = np.random.binomial(1, 0.5, 1000)
Y = X[:, 0] + 2 * D + 0.5 * strata + np.random.randn(1000)

# Fit stratified estimator
estimator = SimpleStratifiedDistributionEstimator()
estimator.fit(X, D, Y, strata)

# Compute QTE at specific quantiles
quantiles = np.array([0.25, 0.5, 0.75])  # 25th, 50th, 75th percentiles
qte, lower, upper = estimator.predict_qte(
    target_treatment_arm=1,
    control_treatment_arm=0,
    quantiles=quantiles,
    n_bootstrap=100
)

print(f"QTE at quantiles {quantiles}: {qte}")
print(f"Median effect (50th percentile): {qte[1]:.3f}")

Source code in dte_adj/base.py 
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
def predict_qte(
    self,
    target_treatment_arm: int,
    control_treatment_arm: int,
    quantiles: Optional[np.ndarray] = None,
    alpha: float = 0.05,
    n_bootstrap=500,
    display_progress: bool = True,
) -> Tuple[np.ndarray, np.ndarray, np.ndarray]:
    """
    Compute Quantile Treatment Effects (QTE) based on the estimator for the distribution function.

    The QTE measures the difference in quantiles between treatment groups, providing insights
    into how treatment affects different parts of the outcome distribution. For stratified
    estimators, the computation properly accounts for strata.

    Variance is estimated by stratified bootstrap: indices are resampled with replacement
    within each stratum independently, which preserves per-stratum sample sizes and reflects
    the covariate-adaptive randomization (CAR) design. For estimators without strata
    (single stratum), this degenerates to a plain bootstrap.

    Args:
        target_treatment_arm (int): The index of the treatment arm of the treatment group.
        control_treatment_arm (int): The index of the treatment arm of the control group.
        quantiles (np.ndarray, optional): Quantiles used for QTE. Defaults to [0.1, 0.2, ..., 0.9].
        alpha (float, optional): Significance level of the confidence bound. Defaults to 0.05.
        n_bootstrap (int, optional): Number of bootstrap samples. Defaults to 500.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        Tuple[np.ndarray, np.ndarray, np.ndarray]: A tuple containing:
            - Expected QTEs (np.ndarray): Treatment effect estimates at each quantile
            - Lower bounds (np.ndarray): Lower confidence interval bounds
            - Upper bounds (np.ndarray): Upper confidence interval bounds

    Example:
        ```python
        import numpy as np
        from dte_adj import SimpleStratifiedDistributionEstimator

        # Generate stratified sample data
        X = np.random.randn(1000, 5)
        strata = np.random.choice([0, 1, 2], size=1000)
        D = np.random.binomial(1, 0.5, 1000)
        Y = X[:, 0] + 2 * D + 0.5 * strata + np.random.randn(1000)

        # Fit stratified estimator
        estimator = SimpleStratifiedDistributionEstimator()
        estimator.fit(X, D, Y, strata)

        # Compute QTE at specific quantiles
        quantiles = np.array([0.25, 0.5, 0.75])  # 25th, 50th, 75th percentiles
        qte, lower, upper = estimator.predict_qte(
            target_treatment_arm=1,
            control_treatment_arm=0,
            quantiles=quantiles,
            n_bootstrap=100
        )

        print(f"QTE at quantiles {quantiles}: {qte}")
        print(f"Median effect (50th percentile): {qte[1]:.3f}")
        ```
    """
    if quantiles is None:
        quantiles = np.arange(1, 10) / 10
    if np.any((quantiles <= 0) | (quantiles >= 1)):
        raise ValueError("quantiles must be in the open interval (0, 1)")

    qte = self._compute_qtes(
        target_treatment_arm,
        control_treatment_arm,
        quantiles,
        self.covariates,
        self.treatment_arms,
        self.outcomes,
        self.strata,
    )

    # Precompute stratum indices for stratified bootstrap.
    # When there is a single stratum this is equivalent to plain bootstrap.
    unique_strata = np.unique(self.strata)
    strata_indices = [np.where(self.strata == s)[0] for s in unique_strata]

    qtes = np.zeros((n_bootstrap, qte.shape[0]))
    bootstrap_iter = range(n_bootstrap)
    if display_progress:
        bootstrap_iter = tqdm(bootstrap_iter, desc="Bootstrap QTE")
    for b in bootstrap_iter:
        bootstrap_indexes = np.concatenate(
            [
                np.random.choice(idx, size=len(idx), replace=True)
                for idx in strata_indices
            ]
        )

        qtes[b] = self._compute_qtes(
            target_treatment_arm,
            control_treatment_arm,
            quantiles,
            self.covariates[bootstrap_indexes],
            self.treatment_arms[bootstrap_indexes],
            self.outcomes[bootstrap_indexes],
            self.strata[bootstrap_indexes],
        )

    qte_var = qtes.var(axis=0)

    qte_lower = qte + norm.ppf(alpha / 2) * np.sqrt(qte_var)
    qte_upper = qte + norm.ppf(1 - alpha / 2) * np.sqrt(qte_var)

    return qte, qte_lower, qte_upper

predict 

predict(
    treatment_arm: int,
    locations: ndarray,
    display_progress: bool = True,
) -> np.ndarray

Compute cumulative distribution values.

Args: treatment_arm (int): The index of the treatment arm. locations (np.ndarray): Scalar values to be used for computing the cumulative distribution. display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

Returns: np.ndarray: Estimated cumulative distribution values for the input.

Source code in dte_adj/base.py 
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
def predict(
    self, treatment_arm: int, locations: np.ndarray, display_progress: bool = True
) -> np.ndarray:
    """
    Compute cumulative distribution values.

    Args:
        treatment_arm (int): The index of the treatment arm.
        locations (np.ndarray): Scalar values to be used for computing the cumulative distribution.
        display_progress (bool, optional): Whether to display a progress bar. Defaults to True.

    Returns:
        np.ndarray: Estimated cumulative distribution values for the input.
    """
    if self.outcomes is None:
        raise ValueError(
            "This estimator has not been trained yet. Please call fit first"
        )

    if treatment_arm not in self.treatment_arms:
        raise ValueError(
            f"This target treatment arm was not included in the training data: {treatment_arm}"
        )

    return self._compute_cumulative_distribution(
        treatment_arm,
        locations,
        self.covariates,
        self.treatment_arms,
        self.outcomes,
        display_progress=display_progress,
    )[0]

fit 

fit(
    covariates: ArrayLike,
    treatment_arms: ArrayLike,
    outcomes: ArrayLike,
) -> AdjustedDistributionEstimator

Set parameters.

Args: covariates: Pre-treatment covariates. treatment_arms: The index of the treatment arm. outcomes: Scalar-valued observed outcome.

Returns: AdjustedDistributionEstimator: The fitted estimator.

Source code in dte_adj/simple.py 
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
def fit(
    self, covariates: ArrayLike, treatment_arms: ArrayLike, outcomes: ArrayLike
) -> AdjustedDistributionEstimator:
    """
    Set parameters.

    Args:
        covariates: Pre-treatment covariates.
        treatment_arms: The index of the treatment arm.
        outcomes: Scalar-valued observed outcome.

    Returns:
        AdjustedDistributionEstimator: The fitted estimator.
    """
    covariates = _convert_to_ndarray(covariates)
    treatment_arms = _convert_to_ndarray(treatment_arms)
    outcomes = _convert_to_ndarray(outcomes)

    if covariates.shape[0] != treatment_arms.shape[0]:
        raise ValueError("The shape of covariates and treatment_arm should be same")

    if covariates.shape[0] != outcomes.shape[0]:
        raise ValueError("The shape of covariates and outcome should be same")

    self.covariates = covariates
    self.treatment_arms = treatment_arms
    self.outcomes = outcomes
    self.strata = np.zeros(len(self.covariates))

    return self

  1. Byambadalai, U., Oka, T., & Yasui, S. (2024). Estimating Distributional Treatment Effects in Randomized Experiments: Machine Learning for Variance Reduction. arXiv preprint arXiv:2407.16037. ↩