Statistics and Probability
CSE · Final · Spring 24
Batch 63
Read the questions as text
Read off the PDF by AI, so it may have mistakes. The PDF is the original.
Daffodil International University
Department of Computer Science and Engineering
Faculty of Science & Information Technology
Semester Final Examination, Spring 2024
Course Code: STA 227, Course Title: Statistics and Probability
Level: 2 Term: 2 Batch: 63
Time: 02:00 Hrs
Marks: 40
Answer ALL Questions
[The figures in the right margin indicate the full marks and corresponding course outcomes. All portions of each question must be answered sequentially.]
1. a) A digital marketing agency manages multiple websites for various clients. They are interested in predicting website traffic to optimize their marketing strategies and resource allocation. They collect data on various factors that may influence website traffic, such as advertising expenditure, social media engagement, search engine ranking, and time of day. They want to build a predictive model that can accurately forecast website traffic based on these factors. [2] CO1
Define a multiple linear regression model to predict website traffic based on various IT-related factors.
1. b) Consider the experiment of rolling a pair of 6-sided dice simultaneously and show the Sample Space. Also Find the probability of same numbers. [2]
1. c) Tell the disadvantage of correlation coefficient 'r'. [1]
2. Suppose a student collected data regarding the height and weight of some students of his section. The data is shown below: [2] CO2
Height (in inches) | 65 | 68 | 70 | 63 | 72 | 66 | 69 | 71
Weight (in kg) | 68 | 72 | 77 | 70 | 82 | 75 | 74 | 79
2. a) Compare which characteristic is showing more variability? [5]
2. b) Show the shape characteristics by the formula of skewness of more consistent variable from the above and comment on it. [5]
3. A large IT firm is interested to find a measure that can be used to fix the wages (yearly) of skilled engineers. On experimental basis, the data on the length of service and their yearly wages (in $00) from a group of 10 randomly selected skilled workers are given below: [3] CO3
Length of service | 11 | 7 | 8 | 9 | 5 | 6 | 10 | 12 | 3 | 4
Yearly wages | 14 | 11 | 13 | 10 | 9 | 10 | 14 | 16 | 6 | 7
3. a) Construct the estimated regression model. [4]
3. b) Identify the Yearly wages of employees who have 15 years Length of service. [1]
3. c) Develop the "Coefficient of determination" of the model. [5]
4. A record of 200 engineers is collected from the HR department of a IT Firm and shown below: [5] CO4
Age/Degree | BSc only | MSc
Below 30 | 90 | 10
30-40 | 20 | 30
Above 40 | 40 | 10
Simplify the following probabilities of selected engineer i) has Msc) ii) Msc and age above 40) iii) BSc and age between 30 to 40) iv) MSc it is given that above 40.v) Age below 30 or MSc)
4. b) Probability that a patient recovers from a rare blood disease is 0.4. if 5 people are known to have this disease, Simplify the probabilities that, [5]
i) Exactly 3 people will survive ii) At least two survive iii) None will survive iv) find the mean and variance of the distribution.
4. c) Justify that there is a significant relationship between Length of service and yearly wages (use data question no-03) used at the 0.05 significance level, interpret your findings. [5]
[Tabulated Value: 2.306].
Formulas
Basic Concepts of Probability
General Rule of Addition
P (A or B) = P (A U B) = P(A) + P(B) - P(A ∩ B)
Special Rule of Addition
P (A or B) = P (A U B) = P(A) + P(B)
Conditional Probability
P(B|A) = P(A and B) / P(A) = P(A∩B) / P(A)
Hypothesis testing
Z test statistics: Z = (X̄ - μ) / (σ / √n)
t test statistics: t = (X̄ - μ) / (s / √n)
Testing the significance of the correlation coefficient
t = r / sqrt((1 - r^2) / (n - 2))
Measure of Dispersion
Range = X_max - X_min
Mean Deviation, M.D = ∑(X_i - X̄)^2 / n
Population variance: σ^2 = ∑(X_i - μ)^2 / N
Population standard deviation: √σ^2
Sample variance: s^2 = ∑(X_i - X̄)^2 / (n - 1)
Sample standard deviation: √s^2
Coefficient of variation for population, C.V = (σ / μ) * 100
Coefficient of variation for sample, C.V = (s / X̄) * 100
Correlation and Regression analysis
Regression Coefficient
β̂_1 = ∑((x_i - x̄)(y_i - ȳ)) / ∑((x_i - x̄)^2)
β̂_0 = ȳ - β̂_1 * x̄
Correlation Coefficient
r = (N∑xy - (∑x)(∑y)) / sqrt([N∑x^2 - (∑x)^2][N∑y^2 - (∑y)^2])
r = ∑((x_i - x̄)(y_i - ȳ)) / sqrt(∑((x_i - x̄)^2) * ∑((y_i - ȳ)^2))
Binomial Distribution
f(x; n, p) = (nCx) * p^x * q^(n-x) for x = 0, 1, 2, ..., n
0; otherwise
Poison Distribution
f(x; λ) = (e^(-λ) * λ^x) / x! for x = 0, 1, 2, ..., ∞
Shape of the distribution
Coefficient of Skewness,
Sk = (3 * (Mean - Median)) / Standard deviation
Kurtosis β_2 = μ_4 / μ_2^2