Unit 2 Probability
4.1 Probability Distributions
Human blood comes in different types. Each person has a specific ABO type (A, B, AB, or O) and Rh factor (positive or negative). Hence, if you are O+, your ABO type is O and your Rh factor is positive. A probability model for blood types in the United States is given:
-
Blood Type A+ A- B+ B- AB+ AB- O+ O- Probability 0.357 0.063 0.085 0.015 0.034 0.006 0.374 0.066 - What is the probability that a randomly chosen person has blood type A?
- What is the probability that two randomly chosen U.S. residents both have type A blood?
- What is the probability that at least one of the two people does not have type A blood?
\(P(A+\text { or } A-)=P(A+)+P(A-)=0.357+0.063=0.420\)
Hence, roughly 42% of residents in the U.S. have ABO-type A blood.
\(\text { P(Person 1 is type A and Person 2 is type A) = }(0.420)(0.420)=0.1764\)
Hence, there is roughly an 18% chance that two randomly selected U.S. residents will both have type A blood.
\( \text {P(at least one not A)}=1-P(\text { both A } )=1-0.1764=0.8236\)
Therefore, in random samples of size two, we expect at least one of the two people not to be type A about 82% of the time.
A probability distribution is what will probably happen, not necessarily what actually happened. Probability distributions are theoretical, not experimental. These distributions are usually given in the form of a graph, table or formula.
A random variable, x, represents a value associated with each outcome of the distribution.
Qualifiers for probability distributions:
- \(\sum P(x)=1\)
- \(0 \leq P(x) \leq 1\) All probabilities are between impossible and certain, inclusive.
Distribution A:
x values | Probability P(x) |
---|---|
2 | 0.4 |
3 | 1.2 |
5 | 0.1 |
Distribution B:
x values | Probability P(x) |
---|---|
2 | 0.4 |
3 | 0.4 |
5 | 0.2 |
Distribution C:
x values | Probability P(x) |
---|---|
2 | 0.1 |
3 | 0.6 |
5 | 0.1 |
Distribution A is not a probability distribution. It does not meet either qualification.
Distribution B is a probability distribution. It meets both qualifications.
Distribution C is not a probability distribution. It does not meet the first qualification.
Two Types of Random Distributions: Discrete and Continuous
Discrete Distributions:
A random variable is discrete when it has a finite or countable number of possible outcomes that can be listed. A discrete distribution lists each possible outcome for x together with the probability of x occurring. The probability distribution is a relative frequency distribution.
Continuous Distributions:
A random variable is continuous when it has an uncountable number of possible outcomes, represented by an interval on a number line.
- Let x represent the number of cars in the PSCC Hardin Valley parking lots.
- Let x represent the height of a random student on PSCC campus.
Creating a Discrete Probability Distribution
- P (Q1 is correct)
= 1/2 = 0.5 - P (Q2 is correct)
= 1/4 = 0.25 - P (Both are Correct) = P (C and C )
= (1/2)(1/4) = 1/8 = .125 - P (C and W)
= (1/2)(3/4) = 3/8 = .375 - P (W and C)
= (1/2)(1/4) = 1/8 = .125 - P (W and W)
= (1/2)(3/4) = 3/8 = .375 - Use the information in a-f to create a probability distribution table:
Let x = number correct
x values Probability P(x) 0 P(W and W) = 0.375
1 P(C and W) + P(W and C) = 0.5
2 P(C and C) = 0.125
- Does our distribution meet both requirements for a probability distribution?
\(\sum P(x)=1\)
Yes, the probabilities add to 1. \(0 \leq P(x) \leq 1\)
Yes, all the probabilities are less than 1 and positive.
y | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
---|---|---|---|---|---|---|---|---|
p(y) | 0.15 | 0.23 | 0.19 | 0.23 | 0.12 | 0.05 | 0.02 | 0.01 |
- What is the probability that a randomly selected family will consist of at least two people?
- What is the probability that a randomly selected family will have two to four members?
- What is the mean family size?
- What is the standard deviation for family size?
\(P(y \geq 2)=1-P(y\lt2)=1-p(1)=0.85\)
\(\mathrm{P}(2 \leq \mathrm{y} \leq 4)=\mathrm{P}(\mathrm{y}=2 \text { or } \mathrm{y}=3 \text { or } \mathrm{y}=4)=\mathrm{p}(2)+\mathrm{p}(3)+\mathrm{p}(4)=0.65\)
\(\mu=3.22\)
\(\sigma \approx \sqrt{2.512} \approx 1.58\)
\(\text { The minimum usual value is } \mu-2 \sigma=3.22-2(1.58) \text { which is } 0.06\)
\(\text { The maximum usual value is } \mu+2 \sigma=3.22+2(1.58) \text { which is } 6.38.\)
So, yes families with 7 or 8 members are unusual.
(The symbol 0+ denotes a positive probability value that is very small. We are not willing to say this is an impossible event. Use P(x=6) = 0 in your calculations.)
x values | Probability P(x) |
---|---|
1 | .02 |
2 | .13 |
3 | .32 |
4 | .39 |
5 | .14 |
6 | 0+ |
If one account is randomly selected, find the following probabilities for the number of phones associated with the account:
- Probability of exactly 3 phones:
P(x = 3) = .32
- Probability of 4 or more phones:
P(x ≥ 4) = P(x = 4) + P(x = 5) + P(x = 6) = .39 + .14 + 0 = .53
- Probability of at least 2 phones:
P(x ≥ 2) = P(x = 2) + P(x = 3) + P(x = 4) + P(x = 5) + P(x = 6) = .13 + .32 + .39 + .14 + 0 = .98
Or use the complement rule: 1 – P(x = 1) = 1 - .02 = .98
- Probability of at most 3 phones:
P(x ≤ 3) = .02 + .13 + .32 = .47
- Probability of no more than 2 phones:
P(x ≤ 2) = .02 + .13 = .15
- Probability of between 2 and 5 phones:
P(2 < x < 5) = P(x = 3) + P(x = 4) = .32 + .39 = .71
- Probability of between 2 and 5 phones, inclusive:
P(2 ≤ x ≤ 5) = P(x = 2) + P(x = 3) + P(x = 4) + P(x = 5) = .13 + .32 + .39 + .14 = .98
Mean (Expected Value) and Standard Deviation of a Discrete Probability Distribution
x values | Probability P(x) |
---|---|
1 | .02 |
2 | .13 |
3 | .32 |
4 | .39 |
5 | .14 |
6 | 0+ |
- Did we keep both rules for probability distributions?
Yes \(\sum P(x)=1\)
\(0 \leq P(x) \leq 1\)
- MEAN (EXPECTED VALUE): \(\mu=E(x)=\sum[x * P(x)]\)
\(\mu=[1 * .02]+[2 * .13]+[3 * .32]+[4 * .39]+[5 * .14]+[6 * 0]=3.5\)
- VARIANCE:
\( \sigma^2=(.95)^2=.9025=.90\)
- STANDARD DEVIATION:
\(\sigma=\sqrt{\sigma^{2}}=\sqrt{.91}=.95\)
- UNUSUAL VALUES:
Min usual value: \(\mu-2 \sigma = \)
\(3.5-2(.95)=1.6\) Max usual value: \(\mu+2 \sigma = \)
\(3.5+2(.95)=5.4\) Would selecting an account with 1 phone be unusual?
Yes, 1 is less than the min usual value. Would selecting an account with 2 phones be unusual?
No, 2 is within the usual range. Would selecting an account with 5 phones be unusual?
No, 5 is within the usual range.