Beta–Binomial Evidence Workbench

How to use Beta–Binomial Evidence Workbench

Add observed successes to alpha and failures to beta. Use the resulting shared-rate posterior to calculate every beta-binomial future count, checking total probability, predictive mean and variance with exact fractions.

1. Use the example as the schema; replace its labels and entered model assumptions.
2. Add observed successes to alpha and failures to beta. Use the resulting shared-rate posterior to calculate every beta-binomial future count, checking total probability, predictive mean and variance with exact fractions.
3. Calculate, inspect the complete witnesses and limitations, then copy or download the JSON report.

Good to know

• Enter positive alpha and beta of at most 10,000, observed success and failure counts from 0 through 100,000 each, and 0–50 future trials. Shapes are exact decimal strings with up to six places; counts are plain integers.
• Assumes one stable Bernoulli rate, conditionally independent observations and future trials given that rate, and an intentionally chosen proper beta prior. Predictive trials share rate uncertainty. No automatic prior choice, credible interval, hypothesis decision, medical interpretation or guarantee about future performance.
• Local calculation with no upload, input persistence or outside lookup. Model JSON is limited to 30,000 characters and 10,000 values. Decimal quantities are strings with up to six places; count fields use plain JSON integers. Exact fractions are capped at 8,192 bits. The complete JSON output is limited to 1,000,000 characters; approximate numbers preserve their stored values. Unsupported states fail explicitly.