import numpy as np

def bootstrap_conversion(data, n_bootstrap=10000):
    bootstrapped_means = []
    for _ in range(n_bootstrap):
        sample = np.random.choice(data, size=len(data), replace=True)
        bootstrapped_means.append(np.mean(sample))
    return bootstrapped_means

**Бета-распределение (аналитически):**  
Если `conversions` – число успехов, `trials` – общее число наблюдений:  
```python
from scipy.stats import beta

conversions_A, trials_A = 100, 1000
conversions_B, trials_B = 120, 1000

beta_A = beta(conversions_A + 1, trials_A - conversions_A + 1)
beta_B = beta(conversions_B + 1, trials_B - conversions_B + 1)