Bayes' Theorem of Probability With Tree Diagrams & Venn Diagrams
Introduction to Bayes' Theorem
Focus on Bayes' theorem and its applications.
Conditional Probability Formulas
Definition: The probability of event A occurring given that event B has occurred is denoted as P(A|B).
Formula: P(A|B) = P(A and B) / P(B)
Reverse Conditional Probability: The probability of event B given A is determined by:
Formula: P(B|A) = P(B and A) / P(A)
Equality of Joint Probabilities: P(A and B) = P(B and A)
Key relationship derived from these formulas:
P(A|B) * P(B) = P(B|A) * P(A)
Bayes' Theorem Derivation
Rearranging the joint probabilities leads to:
P(A|B) = [P(B|A) * P(A)] / P(B)
Usage of Bayes' Theorem:
Helps in calculating conditional probabilities in reverse scenarios.
Example: Probability Calculation Using Bayes' Theorem
Events Defined:
Event A: Drawing numbers 1, 2, 3, 4, and 5.
Event B: Drawing numbers 4, 5, 6, 7, 8, and 9.
Intersection of A and B:
Common numbers: 4 and 5.
Calculations Needed:
P(A) = 5/9 (numbers in A out of total)
P(B) = 6/9 (numbers in B out of total)
P(B|A) = 2/5 (numbers in A that are also in B)
Application of Bayes' Theorem:
Formula:
P(A|B) = [P(B|A) * P(A)] / P(B)
Calculation:
Substitute: P(A|B) = [2/5 * 5/9] / (6/9)
Simplifying results in P(A|B) = 1/3.
Confirmation Using Conditional Probability
Direct Calculation:
P(A and B) = 2/9 (numbers common to A and B)
P(B) = 6/9 (numbers in B)
Thus, P(A|B) = (2/9) / (6/9) = 2/6 = 1/3, confirming Bayes' theorem result.
Another Example: Medical Screening and Prostate Cancer
Problem Statement: Calculate the probability a man has cancer given a positive test result.
Given Information:
Probability of having cancer, P(C) = 12% = 0.12
Positive test result given cancer, P(Pos|C) = 95% = 0.95
Positive test result given no cancer, P(Pos|~C) = 6% = 0.06
Finding P(Pos):
Understanding rates and using a tree diagram:
P(C) = 0.12, P(~C) = 0.88 (1 - 0.12)
Calculate positive outcomes:
P(C and Pos) = P(C) * P(Pos|C) = 0.12 * 0.95 = 0.114
P(~C and Pos) = P(~C) * P(Pos|~C) = 0.88 * 0.06 = 0.0528
Total positive results: P(Pos) = 0.114 + 0.0528 = 0.1668
Final Calculation Using Bayes' Theorem:
Formula: P(C|Pos) = [P(Pos|C) * P(C)] / P(Pos)
Calculation: P(C|Pos) = [0.95 * 0.12] / 0.1668 = 0.68345.
Conclusion: Approximately 68.3% chance of having cancer given a positive test result.
Understanding False Positives
Evaluate the probability a man does not have cancer despite a positive test result:
1 - P(C|Pos) = 100% - 68.3% = 31.7%.
Alternative Calculation Using Population Size
Population scenario analysis: Assuming 10,000 individuals:
With cancer: 1,200, without cancer: 8,800.
Positive outcomes among those with cancer: 1,140 (95% of 1,200).
Positive outcomes among those without cancer: 528 (6% of 8,800).
Total positive results: 1,668 (1,140 + 528).
Result reconfirms previous calculations: P(C|Pos) = 1,140 / 1,668 = 0.683.
Conclusion
Two methods to arrive at the same probability using Bayes' theorem and scenario modeling.
Key takeaway: Understanding conditional probabilities leads to better interpretation of real-world scenarios.