Statistics

ST 361: Introduction to Statistics Introduction to Probability

Kimberly Weems [email protected] 5260 SAS Hall
1

Outline
Probability Trees Probability Models
Sample Spaces, Events, Venn Diagrams

Axioms of Probability Probability Rules (Laws)

Addition Rule Multiplication Rule

Statistics

Example: Southwest Energy

A Southwest Energy Company pipeline has 3 safety shutoff valves in case the line starts to leak. The valves are designed to operate independently of one another:
7% chance that valve 1 will fail 10% chance that valve 2 will fail 5% chance that valve 3 will fail

If there is a leak in the line, find the following probabilities:

a. b. c. d. That all three valves operate correctly That all three valves fail That only one valve operates correctly That at least one valve operates correctly
Statistics

Probability Tree Approach

A probability tree is a useful way to visualize this problem and to find the desired probability.

Statistics

A: P(all three valves operate correctly)

P(all three valves work) = .93*.90*.95 = .79515

Statistics

B: P(all three valves fail)

P(all three valves fail) = .07*.10*.05 = .00035

Statistics

D: P(at least one valve operates correctly)

7 paths P(at least one valve operates correctly = 1 P(no valves operate correctly) = 1 - .00035 = .99965 1 path

Statistics

Example: AIDS Testing

V={person has HIV}; CDC: P(V)=.006 +: test outcome is positive (test indicates HIV present) -: test outcome is negative clinical reliabilities for a new HIV test:
1. If a person has the virus, the test result will be positive with probability .999 2. If a person does not have the virus, the test result will be negative with probability .990
Statistics

Question 1
What is the probability that a randomly selected person will test positive?

Statistics

Probability Tree
clinical reliability

clinical reliability
Statistics

Probability Tree
clinical reliability Multiply branch probs

clinical reliability
Statistics

Question 1 Answer
What is the probability that a randomly selected person will test positive? P(+) = .00599 + .00994 = .01593

Statistics

Question 2
If your test comes back positive, what is the probability that you have HIV? (Remember: we know that if a person has the virus, the test result will be positive with probability .999; if a person does not have the virus, the test result will be negative with probability .990). Looks very reliable

Statistics

Question 2 Answer
Answer two sequences of branches lead to positive test; only 1 sequence represented people who have HIV. P(person has HIV given that test is positive) =.00599/(.00599+.00994) = .376

Statistics

Summary
Question 1: P(+) = .00599 + .00994 = .01593 Question 2: two sequences of branches lead to positive test; only 1 sequence represented people who have HIV. P(person has HIV given that test is positive) =.00599/(.00599+.00994) = .376

Statistics

Recap
We have a test with very high clinical reliabilities:
1. If a person has the virus, the test result will be positive with probability .999 2. If a person does not have the virus, the test result will be negative with probability .990

But we have extremely poor performance when the test is positive: P(person has HIV given that test is positive) =.376 In other words, 62.4% of the positives are false positives! Why? When the characteristic the test is looking for is rare, most positives will be false.
Statistics

Probability models
A probability model is a mathematical representation of a random phenomenon. It is defined by its sample space, events within the sample space, and probabilities associated with each event.

Statistics

Sample Space
Random experiments have unique outcomes. The set of all possible outcome of a random experiment is called the sample space, S. S is discrete if it consists of a finite or countable infinite set of outcomes. S is continuous if it contains an interval (either a finite or infinite width) of real numbers.

Statistics

Example: Sample Spaces

[S is continuous] Randomly select and measure the thickness of a part. S = R+ = {x|x > 0}, the positive real line. Negative or zero thickness is not possible. [S is continuous, finite width] It is known that the thickness is between 10 and 11 mm. We have S = {x|10 < x < 11}. [S is discrete] It is known that the thickness has only three values. S = {low, medium, high}, discrete. [S is discrete] Does the part thickness meet specifications? S = {yes, no}, discrete.

Statistics

19

Example: Sample Spaces (contd)

[S is continuous] Two parts are randomly selected & measured. S = R+ x R+, S is continuous. [S is discrete] Do the 2 parts conform to specifications? S = {yy, yn, ny, nn}, S is discrete. [S is discrete] Number of conforming parts? S = {0, 1, 2}, S is discrete. [S is discrete, countable infinite ] Parts are randomly selected until a non-conforming part is found. S = {n, yn, yyn, yyyn, }, S is countably infinite.

Statistics

20

Events Are Sets of Outcomes

An event (E) is a subset of the sample space of a random experiment, i.e., one or more outcomes of the sample space. Event combinations are: Union of 2 events = the event consisting of all outcomes that are contained in either of two events, E1 U E2. Called E1 or E2. Intersection of 2 events = the event consisting of all outcomes that contained in both of two events, E1 E2. Called E1 and E2. Complement of an event = the set of outcomes that are not contained in the event, E or not E, or Ec .

Statistics

21

Example: Discrete Event Algebra

Recall the sample space from Example 2, S = {yy, yn, ny, nn} concerning conformance to specifications. Let E1 denote the event that at least one part does conform to specifications, E1 = {yy, yn, ny} Let E2 denote the event that no part conforms to specifications, E2 = {nn} Let E3 = , the null or empty set. Let E4 = S, the universal set. Let E5 = {yn, ny, nn}, at least one part does not conform. Then E1 U E5 = S Then E1 E5 = {yn, ny} Then E1 = {nn}
Statistics
22

Example: Continuous Event Algebra

Measurements of the thickness of a part are modeled with the sample space: S = R+. Let E1 = {x|10 x < 12}, show on the real line below. Let E2 = {x|11 < x < 15} Then E1 U E2 = {x|10 x < 15} Then E1 E2 = {x|11 < x < 12} Then E1 = {x|x < 10 or x 12} Then E1 E2 = {x|12 x < 15}

10

11

12

13

14

15

16
23

Statistics

Venn Diagrams Show Event Relationships

Events A & B contain their respective outcomes. The shaded regions indicate the event relation of each diagram.

Statistics

Venn diagrams

24

Venn Diagram of Mutually Exclusive Events

Events A & B are mutually exclusive because they share no common outcomes. The occurrence of one event precludes the occurrence of the other. Symbolically, A B =

Mutually exclusive events

Statistics
25

What is Probability?
Probability is the likelihood or chance that a particular outcome or event from a random experiment will occur. Here, only finite sample spaces ideas apply. Probability is a number in the [0,1] interval. May be expressed as a: proportion (0.15) percent (15%) fraction (3/20) Generally speaking, a probability of: 1 indicates highly likely 0 indicates highly unlikely
Statistics
26

Probability Based on Equally-Likely Outcomes

Whenever a sample space consists of N possible outcomes that are equally likely, the probability of each outcome is 1/N. Example: Consider an unbiased die with 10 faces labeled 1,2,10. Unbiased means that no face is favored, when throwing the die; thus each face has an equal chance of being shown. We throw the die; the probability that the die shows face 10 is 1/10 or 0.1, because each outcome in the sample space is equally likely.

Statistics

27

Probability of an Event
For a discrete sample space, the probability of an event E, denoted by P(E), equals the sum of the probabilities of the outcomes in E. The discrete sample space may be: A finite set of outcomes A countably infinite set of outcomes. Further explanation is necessary to describe probability with respect to continuous sample spaces.

Statistics

28

Equally likely outcomes (uniform model)

Equally likely outcomes (uniform model): If a random phenomenon had k possible outcomes, all equally likely , then each individual outcome has probability 1/k. The probability of any event A is: P(A) = {count of outcomes in A} / {count of outcomes in S} = |A| / k.

Statistics

Example: Probabilities of Events

A random experiment has a sample space {w,x,y,z}. These outcomes are not equally-likely; their probabilities are: 0.1, 0.3, 0.5, 0.1. Event A ={w,x}, event B = {x,y,z}, event C = {z} P(A) = _______ P(B) = _______ P(C) = _______ P(A) = _____ and P(B) = ____ and P(C) = ______ Since event AB = {x}, then P(AB) = _____ Since event AUB = {w,x,y,z}, then P(AUB) = ______ Since event AC = {null}, then P(AC ) = ________
Statistics
30

Axioms of Probability
Probability is a number that is assigned to each member of a collection of events from a random experiment that satisfies the following properties: 1. P(S) = 1 2. 0 P(E) 1, for any event E 3. For each two events E1 and E2 with E1E2 = , P(E1UE2) = P(E1) + P(E2) addition rule These imply that: P() =0 ; P(E) = 1 P(E) complement rule If E1 is contained in E2, then P(E1) P(E2).
Statistics
31

Probability Rules
Joint events are generated by applying basic set operations to individual events, specifically: Unions of events, A U B Intersections of events, A B Complements of events, A

Probabilities of joint events can often be determined from the probabilities of the individual events that comprise it. And conversely.

Statistics

32

Birthday Problem
What is the smallest number of people you need in a group so that the probability of 2 or more people having the same birthday is greater than 1/2? Answer: 23 No. of people 23 30 40 60 Probability .507 .706 .891 .994

Statistics

Birthday Problem
A={at least 2 people in the group have a common birthday} A = {no one has common birthday}
:P ( A') 364 363 365 365

3 people 23 people

: 364 363 343 P ( A') . 498 365 365 365 so P ( A ) 1 P ( A ' ) 1 . 498 . 502
Statistics

Probability rules
Addition Rule for Mutually Exclusive Events: Recall that two events A and B are mutually exclusive (or disjoint) events if they have no outcomes from S in common If A and B are mutually exclusive events, then P(A or B) = P(A) + P(B)

Statistics

Probability rules
General Addition Rule: For any two events A and B

P(A or B) = P(A) + P(B) P(A and B)

Statistics

Probability rules
Multiplication Rule for independent events. Two events, A and B are independent if the occurrence of one does not affect the probability that the other one will occur. If A and B are independent then P(A and B) = P(A) P(B).

Remark on independence: The fact that a coin tossed with my left hand comes up T rather than H, does not influence the outcome of a coin tossed with my right hand. The probability of falling on the street is NOT independent of whether it has snowed. These events are dependent.
Statistics

Some examples
Example1: We roll a 6-sided die. The sample space (set of all possible outcomes) S = ____ The simple events are:___________. The event A that the outcome is {3} is _______. Example 2. We roll a die, and the event of interest, E, is obtaining an odd number. That is E 1,3,5. What is the probability of this event ? Let F 2,4,6 . What is the probability of F?

Statistics