COR10 STATISTICS AND PROBABILITY

REVIEW MATERIAL

DISCRETE RANDOM VARIABLES

• Random Variable – a function that maps outcomes into number; written as a capital letter
• Discrete Random Variable – assumes a countable number of values
• Continuous Random Variable – assumes infinitely large number of values
• Sample Space – total possible outcomes of an experiment
• Event Space – outcome that a certain experiment will occur

Example 1. Let X be a random variable representing the outcome of tossing a fair coin
0, ℎ𝑒𝑎𝑑𝑠
𝑋={
1, 𝑡𝑎𝑖𝑙𝑠
Sample Space: Possible Outcome -> S = {head, tail}
Event Space: E = {head} or E = {tail}

𝑒𝑣𝑒𝑛𝑡 𝑠𝑝𝑎𝑐𝑒
Probability – a possibility that an event will occur ; 𝑃(𝑥) = 𝑠𝑎𝑚𝑝𝑙𝑒 𝑠𝑝𝑎𝑐𝑒

a. What is the probability of getting a face card in ordinary deck or cards?

Event space = face card = (4 Jack, 4 Queen, and 4 King)
Sample space = deck of cards = 52
12 3
𝑃(𝑋 = 𝑓𝑎𝑐𝑒 𝑐𝑎𝑟𝑑) = =
52 13

b. What is the probability of having 2 boy children in a family with 4 children?

𝑆 = 𝐵𝐵𝐵𝐵, 𝐵𝐵𝐵𝐺, 𝐵𝐵𝐺𝐵, 𝐵𝐵𝐺𝐺,
𝐵𝐺𝐵𝐵, 𝐵𝐺𝐵𝐺, 𝐵𝐺𝐺𝐵, 𝐵𝐺𝐺𝐺,
𝐺𝐵𝐵𝐵, 𝐺𝐵𝐵𝐺, 𝐺𝐵𝐺𝐵, 𝐺𝐵𝐺𝐺,
𝐺𝐺𝐵𝐵, 𝐺𝐺𝐵𝐺, 𝐺𝐺𝐺𝐵, 𝐺𝐺𝐺𝐺
Sample space = 16 possible outcomes Event space = BBGG, BGBG, BGGB, GBBG, GBGB, GGBB = 6
6 3
𝑃(𝑋 = 2𝐵) = =
16 8

c. Based on letter b, what is the probability that at least 2 girl children in the family?
Event space = BBGG, BGBG, BGGB, BGGG, GBBG, GBGB, GBGG, GGBB, GGBG, GGGB, GGGG = 11
11
𝑃(𝑋 ≥ 2𝐺) =
16

Probability Mass Function

• ∑𝑃(𝑋 = 𝑥𝑖 ) = 1
• 𝑃(𝑋 = 𝑥) > 0 𝑓𝑜𝑟 𝑥 ∈ 𝑅𝑥
• 0 ≤ 𝑃(𝑋 = 𝑥) ≤ 1 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑥 ∈ 𝑅𝑥

Example 2. A pair of dice is rolled, and the sum is determined.

Sample
Space:

𝑒𝑣𝑒𝑛𝑡 𝑠𝑝𝑎𝑐𝑒
Sum ranges from 2 to 12, that is 𝑅𝑥 = {2,3,4,5,6,7,8,9,10,11,12}; 𝑃(𝑥) = 𝑠𝑎𝑚𝑝𝑙𝑒 𝑠𝑝𝑎𝑐𝑒
 2, 𝑠𝑢𝑚 𝑖𝑠 2 1 6 1 1
3, 𝑠𝑢𝑚 𝑖𝑠 3 𝑃(𝑋 = 2) = 𝑃(𝑋 = 7) = = 𝑃(𝑋 = 12) =
36 36 6 36
4, 𝑠𝑢𝑚 𝑖𝑠 4
5, 𝑠𝑢𝑚 𝑖𝑠 5 2 1 5
𝑃(𝑋 = 3) = = 𝑃(𝑋 = 8) =
6, 𝑠𝑢𝑚 𝑖𝑠 6 36 18 36
𝑋= 7, 𝑠𝑢𝑚 𝑖𝑠 7 3 1 4 1
8, 𝑠𝑢𝑚 𝑖𝑠 8 𝑃(𝑋 = 4) = = 𝑃(𝑋 = 9) = =
36 12 36 9
9, 𝑠𝑢𝑚 𝑖𝑠 9
10, 𝑠𝑢𝑚 𝑖𝑠 10 4 1 3 1
𝑃(𝑋 = 5) = = 𝑃(𝑋 = 10) = =
11, 𝑠𝑢𝑚 𝑖𝑠 11 36 9 36 12
{12, 𝑠𝑢𝑚 𝑖𝑠 12 5 2 1
𝑃(𝑋 = 6) = 𝑃(𝑋 = 11) = =
36 36 18

Probability mass function:

X 2 3 4 5 6 7 8 9 10 11 12

P(X) 1/36 1/18 1/12 1/9 5/36 1/6 5/36 1/9 1/12 1/18 1/36
What is the value of:
1 1 1 1 5
a. 𝑃(𝑋 ≤ 5) = 𝑃(𝑋 = 2) + 𝑃(𝑋 = 3) + 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) = 36 + 18 + 12 + 9 = 18
1 1 5 1 1
b. 𝑃(3 < 𝑋 < 8) = 𝑃(𝑋 = 4) + 𝑃(𝑋 = 5) + 𝑃(𝑋 = 6) + 𝑃(𝑋 = 7) = 12 + 9 + 36 + 6 = 2
1 1 1 1
c. 𝑃(𝑋 > 9) = 𝑃(𝑋 = 10) + 𝑃(𝑋 = 11) + 𝑃(𝑋 = 12) = 12 + 18 + 36 = 6

Example 4. Consider the probability mass function described below. What should be the value of x?
x 1 2 3 4
P(x) 0.15 x 0.30 0.35
Solution: P(X) = 1 (the sum is always equal to 1)
P(X) = 1 – (0.15 + x + 0.30 + 0.35)
x = 1 – 0.15 – 0.30 – 0.35 = 0.20

Expected Value (also known as the mean of X) – average of the probability mass function.
𝜇𝑥 = 𝐸(𝑋) = ∑𝑥 ∙ 𝑃(𝑋 = 𝑥) [multiply the random variable by its probability and then add the result]

Variance - 𝜎 2 = 𝐸[(𝑋 − 𝜇)2 ] = ∑(𝑥 − 𝜇)2 ∙ 𝑃(𝑋 = 𝑥) Standard Deviation - 𝜎 = √𝜎 2 (square-root of variance)

Example 5. Given the probability mass function, solve for the expected value, variance, and standard deviation.
Square the answers in
x P (X = x) 𝑥 ∙ 𝑃(𝑋 = 𝑥) 𝑥−𝜇 4th column
(𝑥 − 𝜇)2
1 1 34 23 23 2 529
1 1/11 1( )= 1− =− (− ) =
11 11 11 11 11 121
2 4 34 12 12 2 144
2 2/11 2( )= 2− =− (− ) =
11 11 11 11 11 121
3 9 34 1 1 2 1
3 3/11 3( )= 3− =− (− ) =
11 11 11 11 11 121
5 20 34 10 10 2 100
4 5/11 4( )= 4− = ( ) =
11 11 11 11 11 121

Expected Value (add the values in 3rd column) Variance (add the values in 5th column)
1 4 9 20 34 529 144 1 100 774
𝜇𝑥 = + + + = 𝜎2 = + + + =
11 11 11 11 11 121 121 121 121 121
The expected value in the sample is 3.09. The variance in the sample is 6.40.

774 3√86
Standard deviation: 𝜎 = √𝜎 2 = √121 = 11
= 2.53 ; 𝜎 = √6.40 = 2.53
NORMAL DISTRIBUTION

Example: Find the area under the normal distribution.

a. 𝑃(𝑍 < −1.26) = 0.1038
Area to the left: 𝑃(𝑍 < 𝑧)
b. 𝑃(𝑍 < −0.72) = 0.2358
• Directly find the area in z-table
c. 𝑃(𝑍 < 2.67) = 0.9962
d. 𝑃(𝑍 < 3.12) = 0.9991 Area to the right: 𝑃(𝑍 > 𝑧)
e. 𝑃(𝑍 > −0.48) = 1 − 0.3156 = 0.6844 • 1 minus the area in z-table

f. 𝑃(𝑍 > 1.16) = 1 − 0.8770 = 0.123 Area between Z: 𝑃(𝑧1 < 𝑍 < 𝑧2 )
g. 𝑃(−1.12 < 𝑍 < −0.58) = 0.2810 − 0.1314 = 0.1496 • find the areas under the z-table
𝑃(𝑍 < −1.12) = 0.1314 𝑃(𝑍 < −0.58) = 0.2810 • subtract smaller value from the
higher value
h. 𝑃(0.89 < 𝑍 < 2.35) = 0.9906 − 0.8133 = 0.1773
𝑃(𝑍 < 0.89) = 0.8133 𝑃(𝑍 < 2.35) = 0.9906

i. 𝑃(−2.59 < 𝑍 < 1.81) = 0.9649 − 0.0048 = 0.9601

𝑃(𝑍 < −2.59) = 0.0048 𝑃(𝑍 < 1.81) = 0.9649

Z-SCORE: a measuring of relative standing

𝑥−𝜇
Formula -> 𝑧 = (z – z-score; x – value of random variable; 𝜇 is population mean; 𝜎 is population standard deviation)
𝜎

Example 5. A factory has a mean mass for chocolate as 50 grams and standard deviation of 2 grams.
a. Find the probability that the chocolate weighs less than 47 grams.
𝑥 − 𝜇 47 − 50 3
𝑧= = = − = −1.5
𝜎 2 2
𝑃(𝑍 < −1.50) = 0.0668 [look for the value of -1.5 using the z-table)

b. Find the probability that the chocolate weighs less than 55 grams.
𝑥 − 𝜇 55 − 50 5
𝑧= = = = 2.5
𝜎 2 2
𝑃(𝑍 < 2.50) = 0.9938 [look for the value of 2.5 using the z-table]

c. Find the probability that the chocolate weights more than 48 grams.
𝑥 − 𝜇 48 − 50 2
𝑧= = = − = −1
𝜎 2 2
𝑃(𝑍 > 1) = 1 − 0.1587 = 0.8413 [look for the value of -1.0 using the z-table]

d. Find the probability that the chocolate weights more than 53 grams.
𝑥 − 𝜇 54 − 50 4
𝑧= = = =2
𝜎 2 2
𝑃(𝑍 > 2) = 1 − .9772 = 0.0228 [look for the value of 2 using the z-table]

e. Find the probability that the chocolate weighs between 46 grams and 52 grams.
46 − 50 4 52 − 50 2
𝑧1 = = − = −2 𝑧2 = = =1
2 2 2 2
𝑃(−2 < 𝑍 < 1) = 0.8413 − 0.0228 = 0.8185 [look for value of -2 and 1 using z-table and subtract]
𝑃(𝑍 < −2) = 0.0228 𝑃(𝑍 < 1) = 0.8413
 f. Chocolate that weighs less than 45 grams or more than 56 grams will be rejected. How many chocolates are
rejected from a sample of 20,000 chocolates?
45 − 50 5
𝑧1 = = − = −2.5 𝑃1 (𝑍 < −2.5) = 0.0062
2 2
56 − 50
𝑧2 = =3 𝑃2 (𝑍 > 3) = 1 − 0.9987 = 0.0013
2

Rejected chocolates = (𝑃1 + 𝑃2 ) ∗ 20,000 = 0.0062 + 0.0013 = 0.0075 ∗ 20000 = 150

g. A content of a water bottle has a mean of 200 mL and a standard deviation of 10 mL. What is the probability
that the content of eight-bottles exceed 1650 mL?
𝑛 = 8; 𝜇 = 200𝑚𝐿 ; 𝜎 = 10; 𝑥 = 1650𝑚𝐿
𝑥 = 𝑛𝜇 = 8 ∙ 200 = 1600 𝜎𝑥 = √𝑛 ∙ 𝜎 = √8 ∙ 10 = 20√2
𝑥 − 𝜇 1600 − 1650
𝑧= = = −1.77
𝜎𝑥 20√2
𝑃(𝑍 > −1.77) = 1 − 0.0384 = 0.9616

SAMPLING DISTRIBUTION
• Population – the totality of items, things, or people under a consideration
• Sample – a subset of the population
• Parameter – a numerical quantity that characterizes a given population; tells something about the population
• Statistic – any numerical quantity that characterizes a given sample; tells something about the whole population
• Sampling distribution – the probability distribution of a sample statistics
• Sample mean – sum of all data divided by the sample size; 𝜇𝑥 = 𝜇
𝜎2
• Sample variance – population variance divided by n: 𝜎𝑥2 = 𝑛
𝜎
• Sample standard deviation - 𝜎𝑥 =
√𝑛

Example 6. A sample of size 12, whose data are 11, 12, 6, 8, 12, 5, 9, 10, 11, 6, 6, 12, is drawn from a population
having a 𝜇 = 10 and standard deviation 𝜎 = 6.

a. What is the sampling error?

11 + 12 + 6 + 8 + 12 + 5 + 9 + 10 + 11 + 6 + 6 + 12
𝑋̅ = =9
12
Sampling Error = 𝑋̅ − 𝜇 = 10 − 9 = 1

b. Determine the standard deviation of the sampling distribution.

Solve for variance first: Standard Deviation: (get the square root of variance)
𝜎 2 62 36
𝜎𝑥2 = = = =3 𝜎 = √𝜎𝑥2 = √3 ≈ 1.73
𝑛 12 12

Example 7. A school has 900 junior high school students. The average heigh of these students is 68 inches with a
standard deviation of 6 inches. Suppose you draw a sample of 50 students, find the mean, variance, and standard
deviation of the sample means that can be derived from the sample. [Given: 𝑛 = 50; 𝜇 = 68; 𝜎 = 6]
a. Mean: 𝜇𝑥 = 𝜇 = 68 𝑖𝑛
𝜎2 62 36
b. Variance: 𝜎𝑥2 = 𝑛
= 50 = 50 = 0.72 𝑖𝑛
𝜎 6
c. Standard Deviation: 𝜎𝑥 = = = 0.85 𝑖𝑛; 𝜎𝑥 = √𝜎𝑥2 = √0.72 = 0.85 𝑖𝑛
√𝑛 √50
Central Limit Theorem

Example 8. Suppose that the blood pressure of a certain population are normally distributed with mean of 125 and SD of
8. A sample is taken from 36 patients, what is the probability that a sample have a blood pressure less than 122?
Given: 𝑥 = 122; 𝜇 = 125; 𝜎 = 8; 𝑛 = 36
𝑥−𝜇 122 − 125
𝑧= 𝜎 = = −2.25 𝑃(𝑍 < −2.25) = 0.0122
8
√𝑛 √36

The probability that the sample has a blood pressure below 122 is 0.0122 or 1.22%.

Example 9. The average exam score of 50 senior high school students in their Statistics exam is 32 with a standard
deviation of 2. What is the probability that a student will have a score higher that 35?
Given: 𝑥 = 35; 𝜇 = 32; 𝜎 = 8; 𝑛 = 50
𝑥 − 𝜇 35 − 32
𝑧= 𝜎 = = 2.65 𝑃(𝑍 > 65) = 1 − 0.996 = 0.004
8
√𝑛 √50
The probability that a student will get a score higher than 35 is 0.004 or 0.4%.

ESTIMATION OF PARAMETERS

Steps in solving confidence interval (using z-table; 𝒏 ≥ 𝟑𝟎)

Z scores for confidence intervals:

90% → 𝑧 = 1.645
95% → 𝑧 = 1.96
99% → 𝑧 = 2.575

Example 10. A random sample of 200 car owns shows that the average speed is 60 km/hr with a standard deviation of 10
km/hr. Construct a 95% confidence interval for the average speed of cars in the Philippines?

Given: 𝑥̅ = 60 𝜎 = 10 𝑛 = 200
𝜎 10
Standard error: 𝑆𝐸 = = ≈ 0.7071 𝑧 = 1.96
√𝑛 √200

Confidence interval: 𝑥̅ ± (𝑧)(𝑆𝐸)

CI: 60 ± (1.96)(0.7071) = 60 ± 1.3859 Confidence interval: 58.6141 ≤ 𝑥 ≤ 61.3859
Thus, we are 95% confident that the average speed of cars is between 58.6141 km/hr to 61.3859 km/hr.
Example 11. The score of a random sample of 100 high school students on a standardized mathematics test gave a mean
of 78 and a standard deviation of 20. Construct a 99% confidence interval of the mean.

Given: 𝑥̅ = 78 𝜎 = 20 𝑛 = 100 𝑧 = 2.575

𝜎 20
Standard error: 𝑆𝐸 = = =2
√𝑛 √100

Confidence Interval: 𝑥̅ ± (𝑧)(𝑆𝐸) 𝐶𝐼: 78 ± (2.575)(2) = 78 ± 5.15

Confidence Interval: 72.85 ≤ 𝑥 ≤ 83.15
The 99% confidence interval of the scores in standardized mathematics test is between 72.85 and 83.15.

Steps in solving confidence interval (using t-table; 𝒏 < 𝟑𝟎)

You can use the confidence level on the t-table as a guide:

Example: Find the values under the t Table

a. 95% confidence level, n = 25
𝛼 = 1 − 0.95 = .05 𝑑𝑓 = 𝑛 − 1 = 25 − 1 = 24
𝑡0.05,24 = 2.064

b. 90% confidence level, n = 12

𝛼 = 1 − 0.90 = .10 𝑑𝑓 = 𝑛 − 1 = 12 − 1 = 11
𝑡0.10,11 = 1.796

c. 99% confidence level, df = 15

𝛼 = 1 − 0.99 = .01 𝑑𝑓 = 15
𝑡0.01,15 = 2.947

d. 98% confidence level, df = 20

𝛼 = 1 − 0.98 = .02 𝑑𝑓 = 20
𝑡0.02,23 = 2.528

Example 12. Compute the 99% confidence interval for the mean number of characters printed in a number of copies of
manuscript when a sample of 15 has a mean of 1.24 characters and a standard deviation of 0.19.
Given: 𝑛 = 15 𝑥̅ = 1.24 𝜎 = 0.19
Confidence level = 99% = 0.99 df = n – 1 = 15 – 1 = 14
𝛼 = 1 − 0.99 = 0.01 𝑡 = 𝑡14,0.01 = 2.977
0.19
Standard Error: ≈ 0.049 Confidence interval: 𝑥̅ ± (𝑡)(𝑆𝐸)
√15

𝐶𝐼: 1.24 ± (2.977)(0.049) = 1.24 ± 0.147 Confidence interval: 1.093 ≤ 𝑥 ≤ 1.387

The 99% confidence interval of the mean is between 1.093 and 1.387.
Example 13. A recent survey conducted by telecom company aims to determine the average length (in minutes) per
month. The researcher in charge surveyed 25 Filipinos at random and found out that the average length of call they
consume per month is 182 with a standard deviation of 18. Estimate a 90% confidence interval of the average minutes of
call per month.

Given: 𝑛 = 25 𝑥̅ = 182 𝜎 = 18
Confidence level = 90% = 0.90 df = n – 1 = 20 – 1 = 19
𝛼 = 1 − 0.90 = 0.10 𝑡 = 𝑡19,0.1 = 1.729
18
Standard Error: ≈ 4.025 Confidence interval: 𝑥̅ ± (𝑡)(𝑆𝐸)
√20

𝐶𝐼: 182 ± (1.729)(4.025) = 182 ± 6.959 Confidence interval: 175.01 ≤ 𝑥 ≤ 188.959

The 90% confidence interval of the average minutes of call per month is between 175.01 and 188.959.

Estimation of Population Proportion

Example 14. XYZ drugstore tried a new dietary supplement on 500 individuals. From the experiment, it was determined
that 290 loss weight. Calculate a 95% confidence interval for the population who would find the supplement effective.

290
Given: 𝑛 = 500 𝑝̂ = 500 = 0.58 Confidence level: 95% z = 1.96

𝑝̂(1−𝑝̂) 0.58(1−0.58)
Standard Error: 𝑆𝐸 = √ =√ = 0.022 Confidence Interval: 𝑝 ± (𝑧)(𝑆𝐸)
𝑛 500

𝐶𝐼: 0.58 ± (1.96)(0.022) = 0.58 ± 0.043 Confidence Interval: 0.537 < 𝑝 < 0.623
Thus, we are 95% confident that the food supplement is effective to about 53.7% to 62.3% of the population.

Example 15. In a random sample of 150 STEM students, 95 passed the 2nd quarter STEM1 exam. Construct a 90%
confidence interval of students who passed the exam.

95
Given: 𝑛 = 150 𝑝̂ = = 0.63 Confidence level: 90% z = 1.645
150

𝑝̂(1−𝑝̂) 0.63(1−0.63)
Standard Error: 𝑆𝐸 = √ 𝑛
=√ 150
≈ 0.039 Confidence Interval: 𝑝 ± (𝑧)(𝑆𝐸)

𝐶𝐼: 0.63 ± (1.645)(0.039) = 0.63 ± 0.064 Confidence Interval: 0.566 < 𝑝 < 0.649
Thus, we are 90% confident that from 56.6% to 64.9% students passed the 2nd quarter STEM1 exam.
Length of Confidence Interval

𝜎 𝑠
Length for 𝑛 ≥ 30: 𝐿 = 2(𝑧)(𝑆𝐸) = 2(𝑧) ( 𝑛) Length for 𝑛 < 30: 𝐿 = 2(𝑡)(𝑆𝐸) = 2(𝑡) ( 𝑛)
√ √

Example 16. Find the length of 95% confidence interval for the mean of normal population with standard deviation of 3
when the sample size is 52.
Given: 𝑛 = 52 𝜎=3 𝑧 = 1.96 (z-table; 95%)
𝜎 3
𝐿 = 2(𝑧) ( 𝑛) = 2(1.96) ( ) = 1.63 The length of the 95% confidence interval for the mean is 1.63.
√ √52

Example 17. Ten basketball players have an average weight of 240 lbs. The sample standard deviation is 25 lbs. Find the
length of 98% confidence interval of the data.
Given: 𝑛 = 10 𝑠 = 25 𝛼 = 1 − .98 = 0.02 𝑑𝑓 = 𝑛 − 1 = 10 − 1 = 9 𝑡 = 2.821
𝑠 25
𝐿 = 2(𝑡) ( 𝑛) = 2(2.821) ( ) = 36.06 The length of the 98% confidence interval of the data is 44.60 lbs.
√ √10

Sample Size Estimation

Example 18. The manager of the school canteen wants to know the daily allowance of the MMCM students. Based on the
previous study, it was determined that the standard deviation of the allowance of the students is Php100. How large a
sample must the manager select is he wants to be 99% confident about the findings where the true mean differs by Php50?
Given: 𝐶𝑉 = 2.575 𝜎 = 100 𝑀𝐸 = 50
2.575 × 100 2
𝑛=( ) = 26.5225 The manager needs to have 27 students.
50

Example 19. 37% of the students in MMCM have cars. How many students should be surveyed to estimate the true
proportion of students who are driving their car in going to school with 90% confidence interval and an error of 5%?
Given: CV = 1.645 ME = 5% = 0.05 𝜎 = √𝑝̂ (1 − 𝑝̂ ) = √0.37(1 − .37) = 0.4828
1.645 × 0.4828 2
𝑛=( ) = 252.35 The required sample size to achieve the desired outcome is 253 students.
0.05