Professional Documents
Culture Documents
2015
3/11/2015
NYU
Spring
Probability
Probability theory
To describe patterns from random events
Definition
A quantitative measure of the chance that an
event will occur
Sample space
The set of all possible outcomes
Proportion
Poor
326
0.25
Fair
452
0.36
Good
488
0.39
1266
1.00
Total
Rules of Probability
Axioms
For an event, E, 0 P(E) 1
The sample space, S, is exhaustive: P(S) = 1
P(not E) = 1 P(E)
For disjoint outcomes, A and B,
P(A or B) = P(A) + P(B)
For independent outcomes, A and B,
P(A and B) = P(A)P(B)
Conditional Probability
The probability of an event B given the
occurrence of an event A
For dependent events
P ( A and B )
P ( B | A)
P ( A)
- Knowledge of the occurrence of A makes the
occurrence of B more likely
- In general, P(B|A) P(B) and P(A|B) P(A) for
dependent events; P(A and B) = P(B|A) x P(A)
For independent events
- P(B|A) = P(B) and P(A|B) = P(A)
- Knowledge of the occurrence of A makes the
occurrence of B no more or less likely; P(A and B)
= P(A) x P(B)
Conditional Probability
Probability Sampling
Recall our first class
Population vs. sample
Parameter vs. statistic
Use statistics from samples to estimate
parameters of populations
From descriptive statistics to inferential
statistics
How to do it?
Purpose
To select a sample that is as representative as
possible of the population
In a way to allow the use of probability theory
to evaluate the generalizations made from the
sample to the population
Probability Sampling
Simple random sample
The most basic probability sampling design
Properties
- Every member of the population has an equal
chance of being chosen
- With the sample size of n, every combination
of n members has an equal chance of being
chosen
Procedure
- Number the members of the population (id
number)
- Pick the random numbers using a random
numbers table
- Find members whose id numbers correspond
to the picked random numbers
Probability Sampling
Example
- Draw a simple random sample of 5 out of a
total of 20
1. Assign id number: 1 2 3 4 5 6 7 9 10 11 17
18 19 20
2. Pick 5 random numbers b/w 1 and 20: 4 5 6
14 20
3. The sample = {4, 5, 6, 14, 20}
Probability Sampling
Systematic random sample
Population size
K
Sample size
Every Kth member in the total population is
chosen for inclusion in the sample after the first
member of the sample is selected at random from
among the first K members in the population
Probability Sampling
Stratified random sample
Dividing the population into subgroups based
on one or more variables of interest and then
Drawing a simple random sample from each of
the subgroups
Proportionate vs. disproportionate
Population
Strata:
race/ethnicity