Week 1 Class 2- Probability Flashcards
What are the 2 basic rules of probability?
The prob of any event occurring is greater that 0 and less than or equal to 1
The sum of probs for all possible outcomes of an activity must =1
What are the two main types of probability
Objective Approach
Subjective Probability
Objective approach includes
Relative Frequency and Classical/ logical method
Relative frequency
P(event) = # of occurrences of the event/ total # of trials or outcomes
Classical/ logical method
using basic knowledge like flipping a coin
Subjective Probability
No logic or past knowledge available
Opinion polls
Mutually exclusive
only one of the events can occur on any one trial
Collectively exhaustive
the list of outcomes include every possible outcome
P( A U B) =
P(A) +P(B) - P(A + B)
Two events are independent if
P(A and B) = P(A) * P(B)
P(AIB) = P(A)
They have no impact on one another
Conditional Probability
Event occurs given another event already occurs
P(BIA) =
P(A and B) / P (A)
divided by what it depends on
Bayes Theorem
Used to incorporate additional information and help create posterior possibilities from original or prior probabilities
Prior possibilities + new info –> bayes process–> posterior prob
Bayes theorem equation
P(A I B) = P( BIA) * P(A) / P(BIA)* P(A) + P(B I A’) * P(A’)
multiply by what is after I
Random Variable
Assigns a real number to every possible event or outcome
Two types of RV
Discrete: finite set of values
Continuous: Infinite set of values
Expected Value
Central tendency of the distribution. compute as the weighted average of the values of the random variable
Finding expected value of a discrete prob distribution
multiply each possible value of RV, Xi, by the probability that outcome will occur, P(Xi), and adding them all together
Variance of a discrete prob distribution
each value of the Rv is subtracted from the expected value, squared, and * the prob of occurence.
binomial
find the prob of a specific # of successes out of n trials
Binomial equation
(n!/ r! (n-r)!) * p^r * q ^n-r
n= trials. p = prob of success in 1 trial. r= # of successes q=1-p
Empiral rule
68% of values are plus or minus 1 st dev from mean
about 95% of values are plus or minus 1 st dev from mean
about 99.7% of values are plus or minus 1 st dev from mean
Poisson dist
involves rate (llamda) = llamda e ^- llamda/ x! llamda = mean rate e= 2.718 x= # of occurences