What is Posterior Probability?

S7-SA3-0437

Grade Level:

Class 12

AI/ML, Physics, Biotechnology, FinTech, EVs, Space Technology, Climate Science, Blockchain, Medicine, Engineering, Law, Economics

Definition

What is it?

Posterior probability is the updated probability of an event happening AFTER we have observed some new evidence or data. It combines our initial belief (prior probability) with the information from the new evidence using Bayes' Theorem.

Simple Example

Quick Example

Imagine you think there's a 30% chance your friend will bring samosas to school today (your initial belief). Then, you see him walking towards school with a large, lumpy bag (new evidence). The updated probability that he has samosas, considering the bag, is the posterior probability.

Worked Example

Step-by-Step

Let's say a factory makes LED bulbs. 2% of all bulbs are faulty. The machine that tests bulbs is not perfect: it correctly identifies a faulty bulb 90% of the time, but it also incorrectly flags a good bulb as faulty 5% of the time.

What is the probability that a bulb is actually faulty, GIVEN that the machine flagged it as faulty?

1. **Define events:**
* F = Bulb is Faulty (P(F) = 0.02)
* G = Bulb is Good (P(G) = 1 - P(F) = 0.98)
* T = Machine Tests as Faulty

2. **Known probabilities (from problem):**
* P(T|F) = Probability of testing faulty GIVEN it's actually faulty = 0.90 (correct detection)
* P(T|G) = Probability of testing faulty GIVEN it's actually good = 0.05 (false positive)

3. **Goal:** Find P(F|T) = Probability of being faulty GIVEN it tested faulty.

4. **Calculate P(T) - Total probability of testing faulty:**
P(T) = P(T|F) * P(F) + P(T|G) * P(G)
P(T) = (0.90 * 0.02) + (0.05 * 0.98)
P(T) = 0.018 + 0.049 = 0.067

5. **Apply Bayes' Theorem:**
P(F|T) = (P(T|F) * P(F)) / P(T)

6. **Substitute values:**
P(F|T) = (0.90 * 0.02) / 0.067
P(F|T) = 0.018 / 0.067
P(F|T) approx 0.2686

**Answer:** The posterior probability that a bulb is actually faulty, given that the machine flagged it as faulty, is approximately 26.86%.

Why It Matters

Posterior probability helps us make smarter decisions by updating our beliefs with new information. It's crucial in AI for making predictions, in medicine for diagnosing diseases based on test results, and in FinTech to assess risks for loans. Understanding this can open doors to careers in data science, medical research, and financial analysis.

Common Mistakes

MISTAKE: Forgetting to include the prior probability in calculations, or treating it as equal for all events. | CORRECTION: Always start with your initial belief (prior probability) for each event before considering new evidence.

MISTAKE: Incorrectly calculating the total probability of the evidence (the denominator in Bayes' Theorem). | CORRECTION: Ensure you sum up all possible ways the evidence could occur, considering both true positives and false positives/negatives.

Practice Questions

Try It Yourself

QUESTION: A rare disease affects 1 in 1000 people. A test for the disease is 99% accurate (it gives a positive result 99% of the time if you have the disease, and a negative result 99% of the time if you don't). If a person tests positive, what is the probability they actually have the disease? | ANSWER: Approximately 9.09%

QUESTION: You hear that a new 'superfood' is advertised to make you score higher marks. 10% of students in your class already score above 90%. After trying the superfood, 30% of students who tried it score above 90%. Also, 5% of students who did NOT try it still score above 90%. If a student scores above 90%, what is the posterior probability that they tried the superfood? (Assume 20% of students tried the superfood). | ANSWER: Approximately 54.55%

QUESTION: A weather app predicts rain with 80% accuracy (if it rains, it predicts rain 80% of the time). If it doesn't rain, it incorrectly predicts rain 10% of the time. In your city, it rains 25% of days. If the app predicts rain today, what is the probability it will actually rain? What is the probability it will NOT rain, given the app predicted rain? | ANSWER: P(Rain|Predicts Rain) = 72.73%; P(No Rain|Predicts Rain) = 27.27%

MCQ

Quick Quiz

Which of the following best describes posterior probability?

The probability of an event before any new evidence is considered.

The probability of an event happening after new evidence has been observed.

The probability of an event that is certain to happen.

The probability of an event that has already occurred.

The Correct Answer Is:

Posterior probability is specifically about updating our belief (probability) after new information or evidence becomes available. Option A describes prior probability.

Real World Connection

In the Real World

In online shopping, when you see 'Recommended for you' products, e-commerce giants like Flipkart use posterior probability. Based on your past purchases (prior belief) and what similar customers bought after viewing certain items (new evidence), they update the probability of you liking a new product. Similarly, medical diagnostic tools use it to refine disease probabilities based on patient symptoms and test results.

Key Vocabulary

Key Terms

PRIOR PROBABILITY: The initial probability of an event before any new information is considered. | LIKELIHOOD: The probability of observing the new evidence, given that a particular hypothesis is true. | EVIDENCE: New data or information that helps update our beliefs. | BAYES' THEOREM: A mathematical formula used to calculate posterior probability, combining prior probability and likelihood. | CONDITIONAL PROBABILITY: The probability of an event occurring given that another event has already occurred.

What's Next

What to Learn Next

Now that you understand posterior probability, you should explore 'Bayes' Theorem' in more detail. It's the core formula for calculating posterior probability and will help you apply this concept to even more complex real-world problems. Keep learning, you're building a strong foundation!