Probability_ MEDC 1114 WEEK 11_31Oct2024 (1)
Week 11: Introduction to Probability Theory
Instructor: Dr. L. Jeyaseelan, Professor of Biostatistics
Objectives
At the end of this tutorial, students should:
Distinguish between probability, proportion at risk, and percentage at risk.
State the laws of probabilities.
Apply operations of probabilities.
Calculate marginal, joint, and conditional probabilities.
Utilize the SPSS statistical package for calculations.
Key Questions Addressed
What is the probability of developing lung cancer for a smoker?
What is the probability of developing diabetes for an obese woman?
Sample Space
Defined as the set of all possible outcomes, denoted as S.
Example: Tuberculin skin test outcomes: S = {Positive, Negative, Uncertain}.
Events
An event is an outcome or a set of outcomes from a random phenomenon.
Example: If a tuberculin skin test shows TB, the event E = {Positive}.
Probability - Definition
Probability quantifies uncertainty and ranges from 0 to 1.
Values:
0 indicates impossibility.
1 indicates certainty.
Classical Definition of Probability
Probability (P) of an event A is given by the formula:
P(A) = (Number of favorable outcomes) / (Total number of possible outcomes)
Example 1: Malnutrition Study
Sample of 50 children aged 5-9 years:
Calculated probability of malnutrition: P(Malnutrition) = 10/50 = 20%.
Sample Space: {Malnutrition, No Malnutrition}.
Empirical/Relative Frequency Probability
For experiments repeated n times with event E occurring m times:
Empirical probability = m/n.
Example 2: Vascular Headache Study
In a clinic, of 300 patients, 30 had vascular headaches:
Probability = P(Vascular Headache) = 30/300 = 0.10.
Axioms (Laws) of Probability
Any probability is between 0 and 1. (0 ≤ p ≤ 1)
Total probability of all possible outcomes must equal 1. (p(S)=1)
Probability of an impossible event is 0.
Complement rule: P(Ac) = 1 − P(A).
Properties of Probability
Certain events have a probability of 1 (e.g., P(Death)=1).
Events that cannot happen have a probability of 0 (e.g., P(Humans are Immortal) = 0).
Laws of Probability
Additive Property (Disjoint Events / Mutually Exclusive)
Events that cannot occur simultaneously (e.g., accidents on Saturday or Sunday):
P(Saturday or Sunday) = P(Saturday) + P(Sunday) = 0.02 + 0.03 = 0.05.
Addition Law of Probability (Not Mutually Exclusive)
For any two events A and B:
Pr(A or B) = Pr(A) + Pr(B) − Pr(A and B).
Example 3: STD Diagnosis
Two doctors diagnosed patients without sharing results:
Diagnoses for STD (Doctor A) = 0.1, Doctor B = 0.17, Overlap = 0.08.
Probability calculation: P(A or B) = P(A) + P(B) − P(A and B) = 0.19.
Multiplication Law of Probability
Probability of two independent events A and B occurring:
P(A and B) = P(A) x P(B).
Example of Independent Events
Considering a couple with two children, probability of both being girls:
P(Girl) = 1/2, P(Girl) for both = (1/2) * (1/2) = 1/4.
Example 4: Probability of Child Characteristics
Probability of a child being male and Rh positive:
P(Male) = 1/2, P(Rh+) = 9/10, P(Single Birth) = 79/80.
Combined probability = (1/2) * (9/10) * (79/80) = 711/1600 = 0.44.
Conditional Probability
Probability of two events occurring together:
Formula: P(A and B) = P(A) * P(B|A) = P(B) * P(A|B).
Example: Probability of lung cancer given smoking status.
Example 5: Lung Cancer Conditional Probability
Disease status probabilities:
Lung cancer (A): Yes = 0.12, No = 0.04 (total = 0.16).
Calculation: P(Lung Cancer | Smoker) = 0.12/0.16 = 0.75.
Marginal Probability
Also known as unconditional probability, not dependent on other events:
Example: Probability of lung cancer in a community.
Joint Probability
Probability of the intersection of two events A and B:
P(A ∩ B).
Example: If diagnosed with lung cancer and smoked, use probabilities to find joint probability.
Practicals and Exercises
SPSS Exercise
Utilize SPSS to reproduce findings from prior slides:
Open SPSS Data: Question 1_HIV and TB.SAV
Open SPSS Data: Question 2_Lung Ca and Smoke.SAV
Questions Relating to Color Blindness
Calculate probabilities associated with gender and color blindness:
P(Color blind) - Marginal probability
P(Male) - Marginal probability
P(Color blind and male) - Joint probability
P(Color blind | male) - Conditional Probability
P(Color blind | woman) - Conditional Probability
Solutions for Color Blindness Questions
P(Color blind) = 15/400 (Marginal)
P(Male) = 200/400 (Marginal)
P(Color blind and male) = 14/400 (Joint)
P(Color blind | male) = 14/200 (Conditional)
P(Color blind | woman) = 1/200 (Conditional)
Assignment
Complete Assignment 5.