Professional Documents
Culture Documents
Sample Space
set of all possible outcomes for an experiment
we like all points in sample space to have same probability
Probabilities of all possible outcomes must add to 1
Toss two coins
HH HT TH TT
Probabilities add to 1 so each point has 1/4 chance
Toss two dice
Event
a subset of the sample space
to which a probability is assigned
Axioms of Probability
Probabilities must satisfy
Probability of any event >=0
Event=All points in sample space
has probability 1
BBG BGB
GBB
3/8
ent approach
e of making
bility change.
probability
1
Law of Complements 2
Events a and Not A are complements 3
Prob(A) + Prob(Not A) = 1 4
A=YELLOW 1 2
Prob(Not A) = 1- Prob(A) B=Not A BLUE
5 6
A and Not A
Law of Complements
Prob(A) + Prob(Not A) = 1
2 3 4 5 6
In general FIRST
Prob(A or B)
=Prob(A) + Prob(B)-Prob( A and B)
Prob(A or B)=
3/12 + 3/12-(1/12)=5/12
If events X and Y are mutually exclusive
P(X and Y)=0
Then Prob( X or Y) = Prob(X) + Prob (Y)
Events A and C are mutually exclusive
so P(A or C)= 3/12 + 2/12 = 5/12
Two events Mutually Exclusive
are mutually exclusive if A=2020 Texans win Super Bowl
the occurrence of one event B=Browns win 2020 Super Bowl
precludes occurrence of the other event Not Mutually Exclusive
A=DOW JONES GOES UP >=20% IN 2020
B=MICROSOFT STOCK GOES UP AT
LEAST 15% IN 2020
SECOND
1 2 3 4 5 6
1
2
3
4 x
5
6
A= card is spade
B= card is ace of spades
A and B are not independent
Machines A B and C
work the following fraction of the time
machine failures independent
A 0.95
B 0.9
C 0.92
Two events A and B are independent if and only if A and B not independent
Prob(A and B) = Prob(A)*Prob(B) A B
A B
A B
A B
A ,B B
A and B independent
A B
13 diamonds A B
13 spades A,B B
13 aces A
13 clubs A
Draw card from deck of cards
A=card is spade
B=card is an ace final IF A, B C, D, ETC ARE INDEPENDENT
A and B are independent
P(A)=1/4 P(B)=1/13 P(A and B) =1/52 PROB(A and B and C and D)=PROB(A
1/52=(1/4)*(1/13) True so A and B are independent
A= card is spade Two independent systems each have 90% chan
B= card is ace of spades What is chance at least one system works?
A and B are not independent
Machines A B and C
P(A) =1/4 work the following fraction of the time
P(B) = 1/52 machine failures independent
P(A and B) =1/52 A 0.95
1/52=(1/4)*(1/52) B 0.9
C 0.92
A and B not independent
What fraction of the time is at least one machine working
0.0004 =0.05*0.1*0.08
Prob(A)=1/2
P(B)= 6/10
Prob(A and B) = 1/10
1/10= (3/5)*(1/2)
Not true so A and B are independent
P(A)=5/20=1/4
P(B)=4/20=1/5
P(A and B)=1/20
1/20=(1/4)*(1/5)
Yes so A and B are independent
are working)
Conditional Probability
P(A and B)
P(A|B)= __________
P(B) CONDITIONAL PROBABILITY: P(A|B) =P(A and B)/P(B)
=1/4
P(A|B)= 1/4
P(A) = 8/20 = 2/5
6
x
x
x
x
x
x
LAW OF TOTAL PROBABILITY P(FL)=.05 P(FAIL|FL)=.8
FIND CHANCE OF AN EVENT P(N FL)=.95 P(FAIL|N FL) = .10
BY ADDING UP ALL THE MUTUALLY EXCLUSIVE WAYS
EVENT CAN HAPPEN
5% cars flood damage 80% of them have engine fail
95% no flood damage only 10% have engine fail
P(Cancer|+)=
Gilbert Welch
Overdiagnosed
Prior Prob*Likelihoods
P(Cancer and +)
----------------
P(+ test result)
Bayes Theorem
Gilbert Welch
States of the World Overdiagnosed
Cancer No Cancer Receive a signal
+Mammogram
- Mammogram
Prior Probabilities
of States of the World Want posterior Probability
before receiving a "signal" P(Cancer|+ test result)=
Prior Prob*Likelihoods
P(Cancer and +)
----------------
P(+ test result)
P(Cancer)*P(+|Cancer)
-----------------------------
P(+ with Cancer)+P(+ with no cancer)
P(Cancer)*P(+|Cancer)
----------------------------
P(Cancer)*P(+|Cancer)+P(No cancer)*P(+|No Cancer)
0.0311284047 =J20/J21
Problem 1 Problem 2
40% of students get an A in marketing We toss two dice
30% of students get an A in statistics A= Event first die is 3
10% get an A in Both B = Event Total is 6
What fraction of students
do not get an A in either course? Are these events independent?
P(No A's ) = 1- P(>=1 A)
P( A and B) = 1/36
P(>=1 A)= P(A in Marketing) +P(A in Stats)- P(A in both) = .4 +.3 -1. P(A) = 1/6
P(B) = 5/36
P(No A's) = 1- .6 = .4 P(A)*P(B) = 5/216
No A and B are not independent
A in Stats No A in Stats
A in Marketing 0.1 0.3 Let C = Event Total is 7
No A in Marketing 0.2 0.4
A and C are independent
P(Event C) = 1/6
P(A and C) =1/36=P(A)*P(Event C)
Problem 3
A bowl has 3 Red and 5 Blue balls Problem 4
We draw 2 balls without replacement 10 coins
What is the chance that we get >=1 Blue Ball 5 Two Headed
3 Two tailed
P(>=1 blue ball) = 1- P( 0 blue balls) 2 Fair
P(0 Blue Balls) =(3/8)*(2/7) = 6/56 What is chance of head?
P(>=1 blue ball) = 1 - 6/56 = 50/56 Given coin toss comes up heads wh
.2*.5
--------------
0.6
Problem 5
P(A)=.1 P(B) =.2 (Event C) = .3
What is P(A or B or C)
If events are Mutually Exclusive?
If events are independent?
Mutually Exclusive
P(A) + P(B) + P(Event C) = 0.6
Independent
P(A or B or C) = 1- P(None of A B and C)
e events independent?
ent Total is 7
C) =1/36=P(A)*P(Event C)
1 P(H|2T) = 0 P(H|F) .5
H) + P(2T)*P(H|2T) + P(F)*P(H|F)
P(F)*P(H|F)
----------------------------------------
1
-------------- = ---------
6