🎲Intro to Probability Unit 12 – Total Probability and Bayes' Theorem

Key Concepts

• Total Probability Theorem calculates the probability of an event by considering all possible ways it can occur
• Bayes' Theorem updates the probability of an event based on new information or evidence
• Conditional probability measures the probability of an event A given that another event B has occurred, denoted as $P(A|B)$
• Mutually exclusive events cannot occur simultaneously; the probability of either event occurring is the sum of their individual probabilities
• Independent events do not influence each other; the probability of both events occurring is the product of their individual probabilities
• Probability distributions describe the likelihood of different outcomes in a random variable
• Law of Total Probability states that the total probability of an event is the sum of its conditional probabilities weighted by the probabilities of the conditioning events

Probability Basics Review

• Probability quantifies the likelihood of an event occurring, ranging from 0 (impossible) to 1 (certain)
• Sample space (S) includes all possible outcomes of a random experiment
• Event (E) is a subset of the sample space; the probability of an event is the sum of the probabilities of its outcomes
• Complement of an event (E') consists of all outcomes in the sample space that are not in the event; $P(E') = 1 - P(E)$
• Union of two events (A ∪ B) contains all outcomes that are in either event A or event B, or both; $P(A ∪ B) = P(A) + P(B) - P(A ∩ B)$
• Intersection of two events (A ∩ B) contains outcomes that are common to both events; for independent events, $P(A ∩ B) = P(A) × P(B)$
• Conditional probability of event A given event B has occurred is $P(A|B) = \frac{P(A ∩ B)}{P(B)}$, where $P(B) > 0$

Total Probability Theorem

• Total Probability Theorem calculates the probability of an event by considering all possible mutually exclusive ways it can occur
• Formula: $P(A) = P(A|B_1)P(B_1) + P(A|B_2)P(B_2) + ... + P(A|B_n)P(B_n)$, where $B_1, B_2, ..., B_n$ are mutually exclusive and exhaustive events
• Requires knowing the conditional probabilities of event A given each of the mutually exclusive events $B_i$
• Partitions the sample space into mutually exclusive events and calculates the probability of A in each partition
• Useful when direct calculation of P(A) is difficult, but conditional probabilities are known or easier to compute
• Example: In a factory, machines A, B, and C produce 50%, 30%, and 20% of the items, respectively. The defect rates for each machine are 2%, 4%, and 3%. The total probability of a randomly selected item being defective is $P(D) = 0.5 × 0.02 + 0.3 × 0.04 + 0.2 × 0.03 = 0.028$

Bayes' Theorem

• Bayes' Theorem updates the probability of an event based on new information or evidence
• Formula: $P(A|B) = \frac{P(B|A)P(A)}{P(B)}$, where P(A) is the prior probability, P(B|A) is the likelihood, and P(B) is the marginal probability
• Relates the conditional probability of event A given event B to the conditional probability of event B given event A
• Requires knowing the prior probabilities of events A and B and the conditional probability of B given A
• Useful in situations where new information becomes available, and the probabilities need to be updated
• Example: In a medical test for a disease with a 1% prevalence, the test has a 95% sensitivity (true positive rate) and a 90% specificity (true negative rate). If a person tests positive, Bayes' Theorem can calculate the probability of actually having the disease: $P(D|+) = \frac{0.95 × 0.01}{0.95 × 0.01 + 0.1 × 0.99} ≈ 0.087$

Applications and Examples

• Bayes' Theorem is widely used in medical diagnosis to update disease probabilities based on test results
• Total Probability Theorem helps calculate the overall probability of an event in complex systems with multiple contributing factors (reliability engineering)
• Conditional probabilities are essential in machine learning and data science for building predictive models and classifiers
• Bayesian inference updates prior beliefs or hypotheses based on observed data, forming the foundation of Bayesian statistics
• Spam filters use Bayesian methods to classify emails based on the presence of certain words or phrases
• In genetics, Bayes' Theorem can calculate the probability of an offspring inheriting a specific trait given the parental genotypes
• Insurance companies use probability theory to assess risk and determine premiums for policyholders

Common Mistakes and Pitfalls

• Confusing conditional probability P(A|B) with joint probability P(A ∩ B) or reversing the order of events
• Failing to ensure that the events in the Total Probability Theorem are mutually exclusive and exhaustive
• Neglecting to update the prior probabilities when applying Bayes' Theorem iteratively
• Misinterpreting the results of Bayes' Theorem, especially when dealing with rare events or low prior probabilities
• Incorrectly assuming that events are independent when they are actually dependent, or vice versa
• Overestimating the accuracy or reliability of probability estimates, especially when based on limited data or subjective judgments
• Falling prey to base rate fallacy by ignoring the prior probabilities and focusing solely on the specific case or evidence

Practice Problems

1. In a factory, 60% of the products are made by machine A, and 40% are made by machine B. Machine A has a 2% defect rate, while machine B has a 5% defect rate. If a randomly selected product is found to be defective, what is the probability that it was made by machine A?
2. A rare disease affects 1 in 10,000 people. A diagnostic test for the disease has a 99% sensitivity (true positive rate) and a 99.9% specificity (true negative rate). If a person tests positive, what is the probability that they actually have the disease?
3. A bag contains 4 red balls and 6 blue balls. Two balls are drawn randomly without replacement. Given that the first ball drawn is red, what is the probability that the second ball is also red?
4. In a certain city, 60% of the taxis are green, and 40% are yellow. On a particular day, 80% of the green taxis and 90% of the yellow taxis are occupied. If a randomly selected taxi is found to be occupied, what is the probability that it is a green taxi?
5. A student answers 70% of the questions correctly when they study and 30% correctly when they don't study. The student studies 60% of the time. What is the probability that the student answers a question correctly?

Real-World Relevance

• Bayes' Theorem is the foundation of Bayesian inference, which has numerous applications in science, engineering, and decision-making under uncertainty
• Total Probability Theorem helps break down complex probability problems into simpler, more manageable components
• Understanding conditional probability is crucial for making informed decisions based on available evidence and data
• Probability theory is essential in risk assessment and management in various fields, such as finance, insurance, and public health
• Machine learning algorithms, such as Naive Bayes classifiers, rely on Bayesian principles to make predictions and classifications
• Probabilistic reasoning is a fundamental skill for critical thinking and problem-solving in everyday life
• Bayes' Theorem has been used in legal cases to update the probability of a defendant's guilt based on evidence presented during a trial