What is Bayes' theorem?

P(A|B) = P(B|A)×P(A) / P(B). It updates a prior belief P(A) with evidence B to get the posterior P(A|B). The likelihood P(B|A) and evidence P(B) drive the update.

Why is base rate important?

Base rate (prior) is how common the condition is. A positive test for a rare disease (1% prevalence) may still mean low probability of having it — base rate neglect leads to overestimating test results.

When do I use simple vs full mode?

Simple: you know P(B) (total probability of evidence). Full: you have P(B|¬A) and compute P(B) via law of total probability: P(B) = P(B|A)P(A) + P(B|¬A)P(¬A).

How do I interpret the posterior?

Posterior P(A|B) is your updated belief after seeing evidence B. Compare to your prior P(A). If posterior >> prior, evidence strongly supports A.

4 more

PROBABILITYProbability TheoryStatistics Calculator

📊

Update Beliefs with Prior, Likelihood, Evidence

From medical diagnosis to spam filters to courtroom evidence — Bayesian reasoning updates beliefs with evidence. Master prior, likelihood, and posterior.

Concept Fundamentals

P(A|B)·P(B)/P(A)

Bayes' Rule

Posterior probability

P(A) — initial belief

Prior

Before evidence

P(B|A)

Likelihood

Evidence given hypothesis

Updated belief

Posterior

After evidence

Calculate PosteriorPrior, Likelihood, Evidence — Step-by-Step Bayesian Reasoning

Why This Statistical Analysis Matters

Why: Beliefs should update when new evidence arrives. Bayes' theorem formalizes this: posterior ∝ likelihood × prior. Base rate neglect and the prosecutor's fallacy are common errors.

How: Enter P(A) prior, P(B|A) likelihood, and P(B) evidence (or P(B|¬A) in full mode). Posterior P(A|B) = P(B|A)×P(A) / P(B).

●P(A|B) ≠ P(B|A) — prosecutor's fallacy
●Base rate matters for rare conditions
●Likelihood ratio measures evidence strength

Sources:Bayes' Original 1763 PaperStanford Encyclopedia - Bayesian Epistemology

📐

BAYESIAN INFERENCEPrior, Likelihood, Evidence — Step-by-Step

Update Beliefs with Prior, Likelihood, Evidence — Step-by-Step Bayesian Reasoning

From medical diagnosis to spam filters to courtroom evidence — Bayesian reasoning updates beliefs with evidence. Master prior, likelihood, and posterior.

Probability Calculator →Normal Distribution →

Calculation Mode

Real-World Scenarios — Click to Load

Inputs

P(A) — Priore.g., disease prevalence

P(B|A) — Likelihoode.g., sensitivity

P(B|¬A) — False positivee.g., 1 - specificity

bayes.sh

CALCULATED

P(A|B) — Posterior

16.10%

Prior P(A)

1.00%

Posterior P(A|B)

16.10%

P(B)

5.9000%

Likelihood Ratio

19.0000

Prior Odds

0.0101

Posterior Odds

0.1919

Bayes Factor

19.0000

Bayesian Inference

P(A|B) — Posterior Probability

16.10%

Prior: 1.0%→Posterior: 16.1%LR: 19.00

numbervibe.com/calculators/statistics/bayes-theorem-calculator

Prior vs Posterior

Posterior Probability

🌳 Probability Tree

Start

A: 1.0%

→ B|A: 95.0% = 0.95%

→ ¬B|A: 5.0% = 0.05%

¬A: 99.0%

→ B|¬A: 5.0% = 4.95%

→ ¬B|¬A: 95.0% = 94.05%

P(B) = P(B|A)P(A) + P(B|¬A)P(¬A) = 0.0095 + 0.0495 = 5.9000%

📊 Interpretation

Low posterior (16.1%): even with positive evidence B, A remains unlikely. This illustrates base rate neglect — a "positive" result doesn't guarantee A when the prior is very low.

1. Prior P(A)

P(A) = 0.0100

2. Likelihood P(B|A)

P(B|A) = 0.9500

3. Evidence P(B)

P(B) = P(B|A)P(A) + P(B|¬A)P(¬A) = 0.009500 + 0.049500 = 0.059000

4. Bayes' Theorem

P(A|B) = \frac{P(B|A) \times P(A)}{P(B)} = \frac{0.009500}{0.059000}

5. Posterior

P(A|B) = 0.1610 (16.10%)

For educational and informational purposes only. Verify with a qualified professional.

📋 Key Takeaways

• Bayes' theorem updates a belief (prior) when new evidence arrives to produce a revised belief (posterior)
• P(A|B) ≠ P(B|A) — confusing these is called the "prosecutor's fallacy" and has led to wrongful convictions
• Base rate neglect: ignoring how rare a condition is leads to overestimating the meaning of positive tests
• The likelihood ratio measures how much stronger evidence is under one hypothesis vs another
• Bayesian thinking is iterative — each posterior becomes the next prior as new evidence arrives

💡 Did You Know

🏛️Rev. Thomas Bayes (1702-1761) never published his theorem — his friend Richard Price found it in his papers after his death

⚖️The prosecutor's fallacy (confusing P(evidence|innocent) with P(innocent|evidence)) has contributed to wrongful convictions worldwide

🤖All modern spam filters, recommendation engines, and language models use Bayesian reasoning at their core

📊During WWII, Alan Turing used Bayesian methods to break the Enigma code, estimating the probability of rotor settings

🧬23andMe and other genetic testing services use Bayes' theorem to calculate your probability of having specific genetic traits

🏥A positive mammogram with 80% sensitivity and 1% prevalence only means ~7.5% chance of actual cancer — base rate matters!

🔍Nate Silver's election models (FiveThirtyEight) are fundamentally Bayesian, continuously updating predictions with new polling data

📖 How It Works

1. Prior Probability

Your initial belief before seeing evidence. P(A) — e.g., disease prevalence in the population.

2. Likelihood

How probable the evidence is if the hypothesis is true. P(B|A) — e.g., test sensitivity.

3. Evidence

Total probability of observing the evidence. P(B) = P(B|A)P(A) + P(B|¬A)P(¬A) via law of total probability.

4. Posterior

Updated belief after seeing evidence. P(A|B) — the answer we seek.

5. Iterative Updating

Each posterior becomes the next prior as new evidence arrives. Bayesian reasoning is inherently sequential.

🎯 Expert Tips

Always consider base rates

A 99% accurate test for a 1-in-10,000 disease still gives mostly false positives.

Use the odds form

Posterior odds = Prior odds × Likelihood ratio is easier for sequential updates.

Beware the prosecutor's fallacy

P(evidence|hypothesis) ≠ P(hypothesis|evidence).

Think about what P(B|¬A) means

The false positive rate dramatically affects the posterior when the prior is low.

⚖️ This Calculator vs. Other Tools

Feature	This Calculator	Probability Calculator	Wolfram Alpha	R	Manual
Dedicated Bayes focus	✅	⚠️ Bayes mode only	✅	⚠️ Manual	✅
Prior vs Posterior charts	✅	❌	❌	❌	❌
Law of total probability mode	✅	❌	⚠️	✅	✅
Likelihood ratio, odds	✅	❌	⚠️	⚠️	✅
Educational content	✅	⚠️	❌	❌	❌
Step-by-step LaTeX	✅	⚠️	⚠️	❌	❌
Example presets	✅	✅	❌	❌	❌
Copy & share	✅	✅	❌	❌	❌

❓ Frequently Asked Questions

What is the difference between Bayes' theorem and regular probability?

Regular probability computes P(A) or P(A and B). Bayes' theorem reverses conditionals: given P(B|A), it finds P(A|B). It's the mathematical framework for updating beliefs with evidence.

Why does the base rate matter so much?

When a condition is rare (low P(A)), even a highly accurate test produces many false positives. P(B|¬A) × (1-P(A)) can dominate the denominator, keeping P(A|B) surprisingly low.

What is the prosecutor's fallacy?

Confusing P(evidence|innocent) with P(innocent|evidence). A 1-in-a-million DNA match doesn't mean 1-in-a-million chance of innocence — it depends on the prior and the size of the suspect pool.

How is Bayes' theorem used in medicine?

Diagnostic tests: given sensitivity P(+|disease), specificity (1 - P(+|healthy)), and prevalence P(disease), Bayes computes P(disease|+). Essential for interpreting mammograms, COVID tests, genetic screening.

What is the difference between prior and posterior?

Prior P(A) is your belief before evidence. Posterior P(A|B) is your updated belief after observing B. Bayes' theorem is the update rule.

What is the likelihood ratio and why does it matter?

LR = P(B|A)/P(B|¬A) measures how much more likely the evidence is under A vs ¬A. LR=10 means evidence is 10× more likely if A is true. Posterior odds = Prior odds × LR.

Can Bayes' theorem be applied iteratively?

Yes. After computing P(A|B₁), use it as the new prior for the next piece of evidence B₂: P(A|B₁,B₂) ∝ P(B₂|A,B₁)P(A|B₁). Each posterior becomes the next prior.

How is Bayesian reasoning used in machine learning?

Naive Bayes classifiers, Bayesian networks, probabilistic graphical models. Modern LLMs use transformer attention that can be interpreted as soft Bayesian inference over context. Spam filters, recommendation systems, and A/B testing all rely on Bayesian updating.

📊 Bayes by the Numbers

1763

Bayes' Paper Published

7.5%

True + from 80% Test, 1% Disease

10⁶

Spam Emails Filtered Daily

P≠P

P(A|B) ≠ P(B|A)

📚 Official Sources

Bayes' Original 1763 Paper ↗

Richard Price's publication of Bayes' posthumous work

Updated: 1763

Stanford Encyclopedia - Bayesian Epistemology ↗

Philosophical foundations of Bayesian reasoning

Updated: 2024

Khan Academy - Bayes' Theorem ↗

Interactive lessons on Bayes' theorem

Updated: 2024

MIT OCW - Probability ↗

Introduction to probability and statistics

Updated: 2022

NIST Engineering Statistics ↗

Probability in engineering applications

Updated: 2023

Wikipedia - Bayes' Theorem ↗

Comprehensive mathematical treatment

Updated: 2024

⚠️ Disclaimer: This calculator provides accurate Bayesian computations for educational and professional reference. For medical diagnosis, legal proceedings, or critical decision-making, consult qualified experts. Base rates and test characteristics vary by population and context.

👈 START HERE

⬅️Jump in and explore the concept!

Update Beliefs with Prior, Likelihood, Evidence

Why This Statistical Analysis Matters

Update Beliefs with Prior, Likelihood, Evidence — Step-by-Step Bayesian Reasoning

Calculation Mode

Real-World Scenarios — Click to Load

Inputs

Prior vs Posterior

Posterior Probability

🌳 Probability Tree

📊 Interpretation

📋 Key Takeaways

💡 Did You Know

📖 How It Works

1. Prior Probability

2. Likelihood

3. Evidence

4. Posterior

5. Iterative Updating

🎯 Expert Tips

Always consider base rates

Use the odds form

Beware the prosecutor's fallacy

Think about what P(B|¬A) means

⚖️ This Calculator vs. Other Tools

❓ Frequently Asked Questions

What is the difference between Bayes' theorem and regular probability?

Why does the base rate matter so much?

What is the prosecutor's fallacy?

How is Bayes' theorem used in medicine?

What is the difference between prior and posterior?

What is the likelihood ratio and why does it matter?

Can Bayes' theorem be applied iteratively?

How is Bayesian reasoning used in machine learning?

📊 Bayes by the Numbers

📚 Official Sources

Related Statistics Calculators

Related Calculators

Bertrand's Box Paradox

Conditional Probability Calculator

Boy or Girl Paradox

False Positive Paradox Calculator

Probability Calculator

Monty Hall Problem Calculator

We Value Your Privacy