Unit 4 Chapter 4 Summary
4.1 Introduction to Random Variables and Probability Distribution
- Discrete Random Variable occurs when the observations of a quantitative random variable can take on only a finite number of values or a countable number of values.
- Continuous Random Variables occurs when the observations of a quantitative random variable can take on any countless numbers of values in a line interval.
Examples of Discrete Random Variables:
- The cost of conducting a medical experiment
- The number of supermodels who ate chocolate cake yesterday
- The number of Business professors who read the Wall Street Journal each day
- The number of Bald Eagles located in New YorkState
Examples of Continuous Random Variables:
- The exact life span (age) of a puppy
- The weight of a feather
- The height of a randomly selected elephant in Africa
- The exact time it takes to calculate 786 – 67
A Probability Distribution is the assignment of probabilities to all specific values of a random variable. A probability is assigned to each value. The probabilities should be between 0 and 1. The sum of all the probabilities must equal 1.
Identify which is a probability distribution.
a.x / P(x)
0 / 0.4219
1 / 0.4219
2 / 0.1406
3 / 0.0156
Sum / 1
This is probability distribution since the values are between 0 and 1 and the sum is 1. / b.
x / P(x)
0 / 0.502
1 / 0.365
2 / 0.098
3 / 0.011
4 / 0.001
Sum / 0.977
This has probabilities between 0 and 1 but the sum is not equal to 1. Therefore b is not a probability distribution.
The mean and standard deviation of a discrete probability distribution are:
Mean
Standard deviation
For the probability distribution a, Excel is used to set up columns and the totals as shown.
x / P(x) / x•P(x) / x - µ / (x - µ)^2 / (x - µ)^2 •P(x)0 / 0.4219 / 0 / -0.7499 / 0.5624 / 0.2373
1 / 0.4219 / 0.4219 / 0.2501 / 0.0626 / 0.0264
2 / 0.1406 / 0.2812 / 1.2501 / 1.5628 / 0.2197
3 / 0.0156 / 0.0468 / 2.2501 / 5.0630 / 0.0790
Sum / 1 / 0.7499 / 0.5623
Mean µ = ∑ (x•P(x)) = 0.7499
Standard Deviation σ = SQRT [ (x - µ)^2 • P(x)] = / 0.7499
Note: It is a coincidence that µ and σ are same in this example.
4.2 Binomial Distribution
Features of a binomial experiment
1. There are afixed number of trials denoted by n.
2. The n trials are independent and performed under identical conditions.
3. Each trial has only two outcomes: success denoted by S and failure denoted by F
4. For each trial the probability of success is the same and denoted by p. The probability of failure is denote by q and p+q=1 (or q = 1 - p)
5. The central problem is to determine the probability of r successes out of n trials. P(r) = ?
Probability for the binomial distribution (probability of r successes out of n trials)
Where r = number of successes
n = number of trials
p = probability of one success
q = probability of one failure
The Table 2 is a simpler way to compute the binomial probability. If you have n, r, and p you can use this table to get P(r). Locate the n group, then the p column. Read the table for r.
For n and r, find the specific probability:
1. n=3, r=0, p=0.05 P(r=0) = 0.857
2.n=15, r=12, p=0.9 P(r=12) = 0.129
3. n=10, r ≤ 4, p=0.6
P(r ≤ 4)= P(0) + P(1) + P(2) + P(3) + P(4)
= .000+.002+.011+.042+.111 = 0.166
4.n=10, r ≥ 8, p = 0.75
P(r ≥ 8) = P(8) + P(9) + P(10)
= .282 + .188 + .056 = 0.526
5.n=5, r is at least 4 (same as 4 and 5), P =0.8
P(at least 4) = P(4) + P(5) = .410 + .328 = 0.738
6.n = 8, r is less than 3 (same as 0,1, 2), p =0.8
P(r < 3) = P(0) + P(1) + P(2) = .000 + .000 + .001 += 0.001
7.The TV called Woops has been successful for many years. The show last season had a share of 20, meaning among the TV sets in use, 20% were tuned to Woops. Assume that an advertiser wants to verify that 20% share value by conducting its own survey. A pilot survey of 10 households having TV sets in use at the same time as the broadcast of Woops. (n = 10, p = 0.2)
- Find the probability that none of the households are tuned to Woops. P(r= 0) = 0.107
- Find the probability that at least one of the households are tuned to Woops. P(at least 1) = 1 – P(0) = 1 - .107 = .893
- Find the probability that at most one of the households are tuned to Woops. P(at most 1) = P(0) + P(1) = .107 + .268 = .375
8. A blind marksman finds that on the average he hits the target 4 out of 5 times. If he fires 4 shots, find the probability of
n = 5, p = 4/5 = 0.8 (hits), q = 1 – 0.8 = 0.2 (misses)
- more than 2 hits
Since problem asks for hits, use 0.8 for the probability.
P(r > 2) = P( r 3 ) = P(3) + P(4) + P(5)
= 0.205 + 0.410 + 0.328 = .943
- At least 3 misses
Since problem asks for misses, use 0.2 for the probability.
P( r 3 ) = P(3) + P(4) + P(5)
= 0.051 + 0.006 + 0.000 = 0 .057
Graphs can be produced for each group and probability on Table 2.
Table 2 for n = 5, p = 0.6r / P (r = )
0 / 0.010
1 / 0.077
2 / 0.230
3 / 0.346
4 / 0.259
5 / 0.078
For a binomial distribution, compute µ and σ .
Mean µ = np n is number of trials
Standard deviation p probability of success
q = 1 – p probability of failure
For n = 5 and p = 0.6
Mean µ = np = 5 ( 0.6) = 3
Standard deviation
Other examples:
n = 12, p = 0.8 q = 1 –p = 1 – 0.8 = 0.2
µ = 12(0.8) = 9.6
Mars, Inc. claims that 20% of its M&M plain candies are orange. A random sample of 100 is selected. Find the mean and standard deviation of the number of orange candies in one such 100 group.
n = 100 p = 0.2 q = 1 – 0.2 = 0.8
µ = 100 (0.2) = 20
- n=5, r=0, p=0.05
P(0) = /
2.n=5, r=2, p=0.05
P(2) =(The slight difference in answer is due to rounding) /
3. n= 5, p = .05, r 5
n= 5, p = .05Find P(r 5) =
P(0) + P(1) + P(2) + P(3) + P(4) + P(5)
= .774 + .204 + .021 + .001 + .000 + .000
= 1 /
4. n = 3, p = .45 r > 1
P(r > 1) = P( r 2 ) = P(2) + P(3) = .334 + .091 .425Or 1 – P( r 1) = 1 – (.166 + .408) .426 (difference due to rounding in table)
5. n = 3, p = .45 , at least 1 for r
P( r 1 ) = P(1) + P(2) + P(3) = .408 + .334 + .091 = .833Examples using the full table
6. n=5, r is at least 4 (same as 4 and 5), P =0.8P(at least 4) = P(4) + P(5) = .410 + .328 = 0.738
7. n=15, r=12, p=0.9 P(r=12) = 0.129
8. n=10, r ≤ 4, p=0.6
P(r ≤ 4)= P(0) + P(1) + P(2) + P(3) + P(4)
= .000+.002+.011+.042+.111 = 0.166
9. n=10, r ≥ 8, p = 0.75
P(r ≥ 8) = P(8) + P(9) + P(10)
= .282 + .188 + .056 = 0.526
10. n=5, r is at least 4 (same as 4 and 5), P =0.8
P(at least 4) = P(4) + P(5) = .410 + .328 = 0.738
11. n = 8, r is less than 3 (same as 0,1, 2), p =0.8
P(r < 3) = P(0) + P(1) + P(2) = .000 + .000 + .001 += 0.001