10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?

method=viewQuestionsInPopup

CHRIST (Deemed to be University),

Questions View-V Semester
Subject : STA531N(LINEAR REGRESSION MODELS)

U1--Simple Linear Regression Model

U1-T1-S1--Introduction

1 Distinguish the relationship between deterministic and probabilistic models. 3.0 Complex(Evaluating/Creating) Descriptive
2 Explain probabilistic relationship in regression model with examples. 3.0 Medium Descriptive
(Applying/Analysing)
3 Explain the role of scatter diagram in regression analysis. 3.0 Simple Descriptive 1
(Remembering/Understanding)
4 What do you mean by regression analysis? Explain its applications. 3.0 Medium Application
(Applying/Analysing)
5 What is scatter diagram? How this diagram is useful in regression analysis? 3.0 Simple Descriptive Parallel 1
(Remembering/Understanding)

U1-T1-S2--Linear Models

6 Prove that the sum of the residuals is zero under simple linear regression. 3.0 Complex(Evaluating/Creating) Application
7 Explain linearity and non-linearity in regression models with examples. 3.0 Medium Descriptive
(Applying/Analysing)
8 What are the assumptions involved in simple linear regression? 3.0 Medium Descriptive
(Applying/Analysing)
9 Explain simple linear regression model with example. 3.0 Simple Descriptive
(Remembering/Understanding)
10 Define coefficient of determination. How its value is interpreted? 3.0 Medium Conceptual
(Applying/Analysing)

U1-T1-S3--Assumptions

11 Explain the sources which attribute the random disturbance "u" in the regression 3.0 Medium Descriptive
models. (Applying/Analysing)
12 Distinguish between the true regression line and estimated regression line. 3.0 Medium Conceptual
(Applying/Analysing)
13 Why the disturbance term is included in the regression model? 3.0 Simple Interpretative
(Remembering/Understanding)
14 How do you find the true values of the population parameters? 3.0 Complex(Evaluating/Creating) Application
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

U1-T2-S1--Applications of SLR

15 2 2 6.0 Complex(Evaluating/Creating) Analytical

Show that , for a straight line fit,r XY = R 2 = r YŶ
16 Show that the square of the multiple correlation coefficient R 2 is equal to the 6.0 Simple Application
(Remembering/Understanding)
square of the correlation coefficient between Y and Ŷ
17 Derive the least square estimator for the regression coefficients in simple linear 6.0 Medium Problem
regression and show that these estimators are unbiased. (Applying/Analysing) oriented
18 Explain the assumptions of simple linear regression model. 6.0 Medium Conceptual
(Applying/Analysing)

U1-T2-S2--Theory of SLR

19 Consider the simple linear regression model y = β 0 + β 1x + ε with 6.0 Medium Descriptive 2
(Applying/Analysing)
E(ε) = 0, Var(ε) = σ 2 and ε uncorrelated. Show that E(MS residuals) = σ 2
20 Explain the procedure to estimate the regression coefficients in 6.0 Simple Descriptive
Y = β 0 + β 1X 1 + ∈ . (Remembering/Understanding)
21 Consider the simple linear regression model y = β 0 + β 1x + ε with 6.0 Complex(Evaluating/Creating) Analytical Parallel 2
2
E(ε) = 0, Var(ε) = σ 2 and ε uncorrelated. Show that E(SS reg) = σ 2 + β 1S xx
22 Explain the method of least square estimation to estimate the regression 6.0 Medium Descriptive
coefficients in Y i = β 0 + β 1X i + ∈ i, (i = 1, 2, …, n). (Applying/Analysing)
23 For simple linear regression model show that mean sum of squares due to 6.0 Medium Conceptual
residuals is an unbiased estimator of variance of error term. (Applying/Analysing)
24 Define coefficient of determination. Show that coefficient of determination is 6.0 Medium Conceptual
equal to the square of correlation coefficient between the observed values of (Applying/Analysing)
independent and dependent variable.
25 Calculate the least square estimates of β 0, β 1, β 2 for the multiple linear regression 6.0 Medium Problem
model (Applying/Analysing) oriented
y = β 0 + β 1x 1 + β 2x 2 + ε
where
Yi={5,4, 2} X1i ={1, 4, 2} and X2i ={2, 1, 0}.

U1-T3-S1--Variances and ANOVA

10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

26 ^ ^ 10.0 Complex(Evaluating/Creating) Analytical

(a) Suppose that we have to fit the straight line regression modely = β 0 + β 1x 1 but the response is
affected by a second variable x 2 such that the true regression function is E(y) = β 0 + β 1x 1 + β 2x 2 . Is the
least squares estimator of the slope in the original simple linear regression model unbiased? If not, show
^
the bias in β 1.
(b) Consider the least squares residuals e i = y i − ŷ, i = 1, 2, . . . n, from the simple linear regression
model. Find the variance of the residuals Var(e i) Is the variance of the residuals a constant? Discuss.
27What is the purpose of testing of the regression coefficients and the overall model? Explain ANOVA 10.0 Simple Descriptive
table for simple linear regression model and derive the test for significance of regression parameters. (Remembering/Understanding)
28 ^ 10.0 Medium Descriptive
(a) Show that Var(β 1) = σ2 in a simple linear regression model Y i = β 0 + β 1X i + ∈ i . Also (Applying/Analysing)
¯
∑ ( Xi − X )2
estimate its standard error.
(b) What is the need of the test of significance of regression coefficients and the whole model? Explain
its procedure under simple linear regression model.
29Derive the expression for variance and standard error of the least square estimate of β 1 in simple linear 10.0 Medium Descriptive
regression model and also construct 99% confidence intervals for β 1. (Assume sample size is large) (Applying/Analysing)

30 Calculate the least square estimates β 0 and β 1 for the simple linear regression model 10.0 Complex(Evaluating/Creating) Problem
oriented
y = β 0 + β 1x + ε
fit to the data
x 0 1 2 3 4
y 1 4 3 8 9

U1-T3-S2--Modeling of SLR

31 Show that, in the general linear regression situation with a β 0 term in the model: 10.0 Medium Conceptual
(Applying/Analysing)
1. The correlation between the vector e and Y is (1 − R 2) 1 / 2.
2. The correlation between e and Ŷ is zero.
32 (a) Explain the procedure to construct 100(1 − α)% confidence interval for regression 10.0 Medium Application
coefficients in Y i = β 0 + β 1X i + ∈ i, (i = 1, 2, . . . . . , n). (Applying/Analysing)
(b) A study was made on the effect of temperature on the yield of a chemical process. A
sample of 11 observations has been taken to perform the analysis and the following table
is the resulting output:
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

Coefficientsa
Unstandardized
Model Coefficients t Pr(>|t|)
Estimate Std. Error
Intercept 9.273 .463 20.021 .000
1
Temperature 1.436 .146 9.807 .000
a. Dependent Variable: chemical process
(i) Identify the least square estimates of the true regression coefficients and form a
regression model
(ii) Estimate the value of chemical process when the temperature is – 4.
(iii) Give 99% confidence interval for both the regression coefficients β 0 and β 1.Interpret
the results.
33 (a) Explain the procedure to construct 95% and 99% confidence intervals for the true 10.0 Simple Application
regression coefficient β 1 in a simple linear regression model (Remembering/Understanding)
Y i = β 0 + β 1X i + ∈ i, (i = 1, 2, . . . . . , n).
(b) The effect of the temperature of the deodorizing process on the color of the finished
product was determined experimentally. A sample of 15 observations collected from the
process and the regression analysis output is as given below:
Coefficients
Unstandardized Standardized
Model Coefficients Coefficients t Pr(>|t|)
Estimate Std. Error Beta
Intercept 2.536 ? 4.802 .000
1
Temperature ? .001 -.726 -3.809 .002
a. Dependent Variable: Color
(i) Estimate the missing values from the above table.
(ii) Form a simple linear regression model.
(iii) Are the regression coefficients are significant? (Use α = 0.05)
(iv) Interpret your result.
34 A test is to be run on a given process for the purpose of determining the effect of an 10.0 Complex(Evaluating/Creating) Application
independent variable X (such as temperature) on a certain characteristic property of the
finished product Y (such as density). Four observations are to be taken at each of five
settings of the independent variable X. When the test was actually run, the following
results were obtained:
Xˉ = 5.0, Yˉ = 3.0, Σ(X − X)
i
ˉ 2 = 160, Σ(Y − Y)
i
ˉ 2 = 83.2, Σ(X − X)(Y
ˉ
i
ˉ = 80.0
− Y)
i
Assume a model of the type Y = β 0 + β 1X + u
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

a. Obtain the fitted regression model.

b. Prepare the analysis of variance table.
c. Determine 95% confidence limits for the true mean value of Y when X=5.0

U2--Multiple Linear Regression Model

U2-T1-S1--Introduction

35 Explain multiple linear regression model with k-explanatory variables using 3.0 Medium Descriptive
matrix notations. (Applying/Analysing)
36 Obtain the normal equations to estimate the regression coefficients of a multiple 3.0 Simple Descriptive
linear regression model with one dependent variable ‘Y’ and two independent (Remembering/Understanding)
variables X1 and X2 using matrix notations.
37 Show that in multiple linear regression model y = Xβ + ε , the least-squares 3.0 Medium Descriptive
estimator can be written as (Applying/Analysing)
β̂ = β + Rε, where R = (X ′ X) − 1X ′ .
38 Consider the multiple linear regression model y = Xβ + ε Show that the least- 3.0 Complex(Evaluating/Creating) Descriptive
squares estimator can be written as
β̂ = β + Rε, where R = (X ′ X) − 1X ′

U2-T1-S2--Applications of MLR

39 Show that the sum of the observed values Yi is equal to the sum of the fitted 3.0 Medium Descriptive
^ (Applying/Analysing)
values Y i.
40 Suppose we fit the model y = X 1β 1 + ε when the true model is actually given by 3.0 Complex(Evaluating/Creating) Descriptive
y = X 1β 1 + X 2β 2 + ε. For both models, assume E(ε) = 0 and Var(ε) = σ 2I. Find
^
the expected value of the ordinary least-squares estimate, β 1. Under what
conditions is this estimate unbiased?
41 Prove that the sum of the residuals weighted by the corresponding fitted value is 3.0 Simple Application
always equals zero. (Remembering/Understanding)
42 n 3.0 Medium Descriptive
Prove that ∑ i = 1ŷ ie i = 0.
(Applying/Analysing)

U2-T1-S3--Estimation
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

43 Show that the error term 'u' follows normal distribution with mean 0 and variance 3.0 Medium Conceptual
σ 2I, where u and 0 are n × 1 vectors, 0 being the null vector and I is n × n identity (Applying/Analysing)
matrix.
44 Show that the regression sum of squares in a linear regression model is 3.0 Complex(Evaluating/Creating) Analytical
n
^2
SSR = ∑ y i − nyˉ 2.
i=1
45 Obtain the mean and variance of the random error "u", where u is the n × 1 vector. 3.0 Medium Problem
(Applying/Analysing) oriented
46 What is the parameter? How does it differ from a variable? 3.0 Simple Conceptual
(Remembering/Understanding)
47 In a multiple linear regression model given by Y= Xβ+ε show that 3.0 Complex(Evaluating/Creating) Conceptual
2
Var(ŷ) = σ uH
where H= X(XTX)-1XT.

U2-T1-S4--Assumptions

48 State and explain Guass Markov theorem. 3.0 Medium Descriptive

(Applying/Analysing)
49 What do you mean by BLUE? Explain. 3.0 Simple Conceptual
(Remembering/Understanding)
50 Find the variance of ŷ. 3.0 Complex(Evaluating/Creating) Analytical
51 Show that the V(ŷ) = σ 2H. 3.0 Medium Conceptual
(Applying/Analysing)

U2-T2-S1--Properties of MLR

52 Prove that SS Res

2 follows a Chi- squared distribution with n-p degrees of freedom.
6.0 Complex(Evaluating/Creating) Descriptive
σ
53 Explain important assumptions of multiple linear regression model. 6.0 Simple Objective
(Remembering/Understanding)
54 ′
Show that the V (b ) = σ 2 (X X ) − 1, where b is the OLS estimator of regression 6.0 Medium Descriptive
(Applying/Analysing)
coefficients.
55 Consider a model y n × 1 = x n × k + 1β k + 1 × 1 + u n × 1 and obtain the variance of b, 6.0 Medium Descriptive
(Applying/Analysing)
where b is an unbiased estimator of β.

U2-T2-S2--Properties of Regression Coefficients

10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

56 Explain OLS method to estimate the regression coefficients in multiple linear 6.0 Simple Descriptive
regression model. (Remembering/Understanding)
57 Give expression for the ordinary least square estimate of β vector in MLR model 6.0 Complex(Evaluating/Creating) Descriptive
and show that these estimates are unbiased.
58 Explain multiple regression models and show that the least square estimates of 6.0 Medium Descriptive
regression coefficients are unbiased. (Applying/Analysing)
59 Explain ordinary least squares estimation method for a multiple linear regression 6.0 Medium Descriptive
model Y = X β + ℮, where Y is (n x1), X is (n x k), β is (nx1) and ℮ is an (nx1) (Applying/Analysing)
order matrices.
60 Let Y= Xβ+ ε, where X has full rank matrix and ε is distributed as Nn(0, σ2I), 6.0 Complex(Evaluating/Creating) Conceptual
then show that the least square estimator
β̂ = (XTX)-1XTY
is distributed as Nk+1(β, σ2 (XTX)-1).
61 For multiple linear regression model Y= Xβ+ε derive the least square estimates of 6.0 Medium Conceptual
β. Also prove that the least square estimators are unbiased estimators of β. (Applying/Analysing)

U2-T2-S3--Estimation of regression coefficients

62 2 6.0
Simple Problem
Obtain an unbiased estimator of σ u.
(Remembering/Understanding) oriented
63 ^ 6.0 Medium Analytical
2 ′
Show that σ u = E n( e− ke ) , where n is the number of observations and k are the number (Applying/Analysing)

of parameters to be estimated in the regression model.

64 2 6.0 Complex(Evaluating/Creating) Analytical
Derive the expression for the estimate of σ in multiple linear regression model.
u
65 2 6.0 Medium Descriptive
Derive the expression of an unbiased estimator of σ u.
(Applying/Analysing)
66 The following table gives the result of ANOVA conducted in regression analysis: 6.0 Complex(Evaluating/Creating) Conceptual
ANOVA
Sum of Mean Sum
Source df F value P(>F)
Squares of Squares
Regression ? 103.87 ?
? 0.0001
Error 16 ? ?
Total 19 123.86
(a) Find the missing values.
(b) Determine
(i) total number of observations
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

(ii) number of independent variables

(iii)the estimated value of variance of error.
(c) How much variation in dependent variables is explained by the regression
model.
(d) What can be concluded regarding the overall significance of fitted regression
model?

U2-T3-S1--Significance and Applications of MLR1

67 (a) Estimate the variance of the least square estimates of regression coefficients in 10.0 Simple Problem
MLR model. (Remembering/Understanding) oriented
(b) Let Y 1 = θ + ε 1, Y 2 = 2θ − θ 1 + ε 2 and Y 3 = θ + 2θ 1 + ε 3. Where
E(ε i) = 0, i = 1, 2, 3. Find the least square estimates of θ and θ 1 .
68 (a) Estimate the variance of the least square estimates of regression coefficients in 10.0 Medium Problem
MLR model. (Applying/Analysing) oriented
2
(b) Consider the regression model E(Y i) = β 0 + β 1X i + β 2(X i − 1)(i = 1, 2, 3). Where
X1= - 1, X2 = 0 and X3 = +1. Find the least square estimates of β 0 , β 1 and β 2.
69 2 10.0 Complex(Evaluating/Creating) Problem
Consider the regression model E(Y i) = β 0 + β 1X i + β 2(X i − 1)(i = 1, 2, 3). Where
oriented
X1= - 1, X2 = 0 and X3 = +1. Find the least square estimates of β 0, β 1 and β 2 and the
residuals associated with them.
70 Let Y i = β 0 + β 1X 1i + β 2X 2i + ε i, (i = 1, 2, 3) , where Y i = {5, 3, 4}, X 1i = {1, 4, 2} 10.0 Medium Problem
and X 2i = {2, 1, 0} respectively. Find the least square estimates of β 0 , β 1 and β 2. (Applying/Analysing) oriented

U2-T3-S2--Significance and Applications of MLR2

71 (a) Explain the test procedure for testing the significance of regression coefficients in MLR 10.0 Medium Problem
model. (Applying/Analysing) oriented
(b) A study was conducted to predict the BMI of persons based on several factors like
Education, Calories, Income and Expenditure of persons. Data were collected and the output
is given below:
Coefficientsa
Unstandardized Standardized
Model Coefficients Coefficients t p-value
Estimate Std. Error Beta
1 Intercept 20.693 .208 ? .000
calorie .002 .000 .753 38.969 .000
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

expenditure ? .003 -.163 -8.434 .000

income 8.819E-005 .000 .035 1.837 .067
education -.001 ? -.002 -.086 .932
a. Dependent Variable: BMI
(i) Fill the missing values (?).
(ii) From the above results form a fitted multiple linear regression model.
(iii) Identify the variables which are having significant influence on BMI.
(iv) Give your conclusion based on Beta values.
(v) Give your conclusion from this output.
72 (a) Prove that the variance of the ordinary least squares estimates is lowest. 10.0 Medium Problem
(b) A study was conducted to know the relationship between number of credit cards, family (Applying/Analysing) oriented
size and monthly income of 20 households in a locality. The output shown below:
ANOVAa
Sum of Mean
Model df F p-value
Squares Square
Regression 35.689 2 ? 9.627 .002b
1 Residual 31.511 17 1.854
Total ? 19
(i) Fill the missing values. (?)
(ii) Are the regression coefficients significant? (use 5% LOS)
(iii) What percent of variation is explained by the model?
73 (a) Show that the variance of the OLS estimates is minimum. 10.0 Simple Problem
(b) A study was conducted and the research identified that BMI is depending on the other (Remembering/Understanding) oriented
factors like Education, Calories, Income and Expenditure of persons. Data were collected
and the output is given below:
ANOVAa
Sum of
Model df Mean Square F p-value
Squares
Regression 1062.377 4 265.594 ? .000b
1 Residual ? 995 .594
Total 1653.771 999
a. Dependent Variable: BMI
b. Predictors: (Constant), education, calorie, income, expenditure
(i) Find the missing values (?).
(ii) Identify how many observations used in this study.
(iii) Estimate the value of σ 2.
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

(iv) Obtain coefficient of determination and interpret it.

74 (a) Explain the test procedure for testing the significance of complete regression in MLR 10.0 Complex(Evaluating/Creating) Problem
model. oriented
(b) A study was conducted to know the relationship between number of credit cards, family
size and monthly income of 20 households in a locality. The output shown below:
Coefficientsa
Unstandardized Standardized
Model Coefficients Coefficients t p-value
Estimate Std. Error Beta
Intercept -1.157 ? -.624 .541
1 F_size .013 .254 .009 .053 .959
F_income .078 .019 .732 ? .001
a. Dependent Variable: Credit_cards
(i) Fill the missing values (?).
(ii) Form a fitted multiple linear regression model.
(iii) Give your conclusion based on Beta coefficients.
(iv) Is this model is suitable to predict the number of credit cards of households? Explain.
75 Following tables gives the results of analysis conducted to fit the regression model 10.0 Complex(Evaluating/Creating) Problem
Y = β 0 + β 1X 1 + β 2X 2 + ∈ oriented
where
Y denotes the selling price of house in thousands of dollars
X1 denotes the total dwelling space in hundreds of square feet
X2 denotes assessed value in thousands of dollars
ANOVA
Sum of Mean Sum
Source df F value P(>F)
Squares of Squares
Regression 2 1032.87 516.43
42.858 0.0001
Error 17 204.99 12.05
Parameter Estimates
Parameter Standard
Variable t P(> t )
Estimates Error
Intercept 30.96 7.88 3.929 0.0011
x1 2.63 0.78 3.353 0.0038
x2 0.04 0.28 0.158 0.8760
(a) Write the regression model and interpret the values of regression coefficients.
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

(b) Test the significance of overall model and individual regression coefficients.(Use 5%
level of significance)

U3--Criteria for Model Selection and Residual analysis

U3-T1-S1--Model Selection

76 ¯2 3.0 Complex(Evaluating/Creating) Conceptual

Explain the meaning and advantage of R criteria. Show that the model with the
¯2
maximum R is also the model with minimum residual mean square.
77 Explain the importance of R2 in regression models. 3.0 Simple Application
(Remembering/Understanding)
78 Explain the role of adjusted R2 as a model selection criterion in regression 3.0 Medium Descriptive
modeling. (Applying/Analysing)
79 What do you mean by coefficient of determination? Distinguish between R 2 and 3.0 Medium Descriptive
¯2 (Applying/Analysing)
R .

U3-T1-S2--Variable Selection

80 Explain step-wise regression method in regression model building. 3.0

Complex(Evaluating/Creating) Conceptual
81 Explain forward method in selecting significant variables in regression modeling. 3.0
Simple Descriptive
(Remembering/Understanding)
82 Explain backward elimination method in regression modeling. 3.0 Medium Descriptive
(Applying/Analysing)
83 Explain the forward selection method of variable selection in regression modeling 3.0 Medium Conceptual
with an example. (Applying/Analysing)

U3-T1-S3--Goodness of fit criterion

84 Explain residual mean square criterion and derive the expression when 3.0 Medium Problem
minimum residual mean square is equal to the maximum adjusted R square. (Applying/Analysing) oriented
85 Explain the need of Mallow's Cp statistic in evaluating the subset regression 3.0 Complex(Evaluating/Creating) Interpretative
model.
86 What are the methods of goodness of fit? Explain Residual Mean Square Error 3.0 Medium Analytical
criterion. (Applying/Analysing)
87 Explain the use of Residual Mean Square Criterion of goodness of fit. 3.0 Simple Descriptive
(Remembering/Understanding)
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

U3-T2-S1--Variable transformation

88 Prove that 6.0 Complex(Evaluating/Creating) Application

¯ ^ ^ ¯
n n n
(a) ∑ i = 1(Y i − Y) = ∑ i = 1(Y i − Y i) + ∑ i = 1(Y i − Y) 2
2 2

(b) R 2 = 1 − SS Res
SS T
89 The following table shows the regression output, with some numbers erased, when a 6.0 Medium Application
simple regression model relating a response variable Y to a predictor variable X is fitted (Applying/Analysing)
based on 18 observations. Complete the missing numbers.
Coefficients Table
Variable Coefficients S.E. t-test p-value
Intercept 3.43179 - 0.265 0.7941
X 0.4576 0.1421 - <0.0001
R 2=0.716 Adjusted R 2= -
90 The table below shows the regression output, with some missing values, where Y is 6.0 Medium Descriptive
regressed on X with 18 observations. Complete the missing numbers and comment on it. (Applying/Analysing)
Coefficients Table
Variable Coefficients S.E. t-test p-value
Intercept 3.43179 - 0.265 0.7941
X 0.4576 0.1421 - <0.0001
R 2=0.716 Adjusted R 2= -
91 Explain the need of transformation of variables in regression analysis and discuss the 6.0 Simple Conceptual
Box-Cox transformation. (Remembering/Understanding)

U3-T3-S1--types of residuals and residual plots

92 What is the need of residual analysis? Explain the Studentized and PRESS 10.0 Complex(Evaluating/Creating) Descriptive
residuals.
93 What do you mean by the term "Residual" in regression analysis? What are its 10.0 Simple Analytical
assumptions? Explain any two types of residuals. (Remembering/Understanding)
94 Define Residuals. What are the types of residuals? Explain Standardized and 10.0 Medium Problem
Studentized residuals. (Applying/Analysing) oriented
95 Define the term "Residual". Explain various types of residual plots in checking 10.0 Medium Interpretative
the normality. (Applying/Analysing)
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

U4--Tests of Assumptions in MLR

U4-T1-S1--Model assumptions-heteroscedasticity

96 Explain the consequences of heteroscedasticity and give any two remedial 3.0 Complex(Evaluating/Creating) Application
measures.
97 If the assumption of homoscedasticity is not satisfied in regression modeling, 3.0 Medium Application
what are the possible consequences? Explain in brief. (Applying/Analysing)
98 Explain the sources and consequences of heteroscedasticity. 3.0 Medium Conceptual
(Applying/Analysing)
99 Explain the concept of homoscedasticity in regression analysis. 3.0 Simple Conceptual
(Remembering/Understanding)

U4-T1-S2--Model assumptions-Multicollinearity

100 Explain the concept of multicollinearity in regression analysis. 3.0 Simple Conceptual
(Remembering/Understanding)
101 Explain the term "variance inflation factor". How do we detect multicollinearity 3.0 Complex(Evaluating/Creating) Application
using VIF?
102 If the coefficient of determination is 0.65, then what would be the Variance 3.0 Medium Problem
inflation factor? (Applying/Analysing) oriented
103 What are the consequences of multicollinearity in regression analysis? 3.0 Medium Conceptual
(Applying/Analysing)

U4-T2-S1--Autocorrelation

104 (a) Define autocorrelation. Explain the consequences of autocorrelation. 10.0 Medium Interpretative
(b) A study was conducted to investigate the relationship between monthly income and (Applying/Analysing)
expenditure of 30 households and the following result obtained:
Model Summaryb
R Adjusted R Std. Error of the Durbin-
Model R
Square Square Estimate Watson
1 .931a .866 .861 5.19203 2.060
Using above table give some useful conclusions about the model assumptions.
105 (i) Define Auto-correlation. What are its consequences? 10.0 Simple Interpretative
(ii) Explain the use of DW test in testing auto-correlation. (Remembering/Understanding)
106 (a)What is Auto-correlation? What happens if there is auto-correlation? 10.0 Medium Application
(Applying/Analysing)
10/11/23, 1:33 AM kp.christuniversity.in/KnowledgePro/syllabuseCreation.do?method=viewQuestionsInPopup

(b)Describe how the problem of auto-correlation may arise. Explain Durbin-Watson

test and conclude on the nature of auto-correlation present.
107 (a)Define autocorrelation. 10.0 Complex(Evaluating/Creating) Problem
(b) Explain Durbin-Watson test. A study was conducted to investigate the relationship oriented
between monthly income and expenditure of 30 households and the following result
obtained:
Model Summaryb
R Adjusted R Std. Error of the Durbin-
Model R
Square Square Estimate Watson
1 .931 a .866 .861 5.19203 2.060
Using above table give some useful conclusions about the model assumptions.