07_weighted_data.rst

Weighted Data

Fit distributions to weighted observations.

Basic Usage

from distfit_pro.core.weighted import WeightedFitting
from distfit_pro import get_distribution
import numpy as np

data = np.random.normal(5, 2, 1000)
weights = np.random.uniform(0.5, 1.5, 1000)

dist = get_distribution('normal')
params = WeightedFitting.fit_weighted_mle(data, weights, dist)
dist.params = params
dist.fitted = True

Weighted Statistics

stats = WeightedFitting.weighted_stats(data, weights)
print(f"Weighted mean: {stats['mean']:.2f}")
print(f"Weighted std: {stats['std']:.2f}")

Effective Sample Size

ess = WeightedFitting.effective_sample_size(weights)
print(f"ESS: {ess:.1f} (out of {len(weights)})")