MATH 402 - LINEAR REGRESSION -
Inference in Linear Regression
Question Bank - Set 2
Liberty University
Question 1
Question
Suppose we have a linear regression model given by: Yi=β0+β1Xi+ϵi, where
ϵi∼ N(0, σ2)and β0, β1are the intercept and slope parameters, respectively.
Given a sample of size n= 20, we obtain the following results:
Variable Estimate
β03.2
β11.5
σ24.0
Test the hypothesis H0:β1= 0 vs. H1:β1= 0 at the 5% significance level.
Solution
Step 1: Define the hypothesis test. Our null hypothesis is H0:β1= 0 and the
alternative hypothesis is H1:β1= 0. We will conduct an F-test to test the
significance of the slope coefficient β1.
Step 2: Calculate the test statistic. The test statistic is given by:
F=SSR/1
SSE/(n−2)
Where SSR is the regression sum of squares, SSE is the error sum of squares,
and nis the sample size.
Step 3: Determine the degrees of freedom. Since we have two parameters
estimated in the model (β0and β1), the degrees of freedom for the regression is
2. The error degrees of freedom is n−2 = 20 −2 = 18.
Step 4: Calculate the test statistic value. We have SSR =β2
1×SSXX and
SSE =SSE/df where SSXX is the sum of squares of Xand df is the degrees of
freedom for the error.
Plugging in the given values, we find:
SSR = 1.52×SSXX
SSE = 4.0/18
Step 5: Determine the critical value. Look up the critical value from the
F-distribution table with degrees of freedom 1 and 18 at the 5% significance
level. Let’s say the critical value is Fcrit.
Step 6: Make a decision. If F > Fcrit, reject the null hypothesis. Otherwise,
fail to reject the null hypothesis.
Step 7: Conclusion. State your conclusion based on the decision made in
step 6.
Question 2
Question
Suppose we have a dataset with the following linear regression model:
Y= 3X1−2X2+ 5X3+ϵ
Where X1, X2, and X3are independent variables, and ϵis the error term. Af-
ter performing linear regression analysis, we obtained the following estimated
regression coefficients:
ˆ
β1= 2.6,ˆ
β2=−1.8,ˆ
β3= 4.9
Also, the standard errors of these estimates were found to be:
SE(ˆ
β1) = 0.3, SE(ˆ
β2) = 0.4, SE(ˆ
β3) = 0.2
Using a significance level of 0.05, test the null hypothesis that the coefficient for
X2is equal to zero.
Solution
Step 1: Formulate the hypotheses.
The null hypothesis (H0) and alternative hypothesis (H1) can be formulated as:
H0:β2= 0
H1:β2= 0
2
Step 2: Calculate the test statistic.
The test statistic for testing the significance of β2is given by:
t=ˆ
β2
SE(ˆ
β2)
Substitute the given values to find the test statistic:
t=−1.8
0.4=−4.5
Step 3: Determine the degrees of freedom.
The degrees of freedom for the t-distribution in this case is n−3, where nis the
number of observations. Since the degrees of freedom is not explicitly given in
the question, we assume that there are enough observations to ensure that the
t-distribution is a good approximation.
Step 4: Determine the critical value.
Using a significance level of 0.05 and the two-tailed test, the critical value is
found using a t-table with appropriate degrees of freedom.
Step 5: Make a decision.
Compare the absolute value of the test statistic with the critical value from the
t-table. If the absolute value of the test statistic is greater than the critical value,
we reject the null hypothesis; otherwise, we fail to reject the null hypothesis.
Step 6: Conclusion.
Based on the decision in Step 5, make a conclusion regarding the null hypothesis
at the 0.05 significance level.
Question 4
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+ϵi, where
ϵi∼N(0, σ2)represents the error term. We are interested in testing the null
hypothesis H0:β1= 0 against the alternative hypothesis H1:β1= 0. Given
the following data and output summary from a regression analysis:
Variable ˆ
βSE(ˆ
β)
β03.56 0.42
β10.78 0.06
Perform the hypothesis test using a significance level of 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing whether the
slope is significantly different from zero is given by:
t=ˆ
β1−0
SE(ˆ
β1)
3
Given ˆ
β1= 0.78 and SE(ˆ
β1) = 0.06, we have:
t=0.78 −0
0.06 = 13
Step 2: Determine the critical value. Since we are testing for a two-tailed
hypothesis test at a significance level of 0.05, the critical value can be obtained
using a t-distribution with n−2degrees of freedom.
For a two-tailed test with α= 0.05 and n−2 = 10 −2=8degrees of
freedom, the critical values are ±2.306.
Step 3: Make a decision. Since |t|= 13 >2.306, we reject the null hypothesis
H0:β1= 0 in favor of the alternative hypothesis H1:β1= 0 at the 0.05
significance level. This indicates that the slope coefficient β1is significantly
different from zero.
Question 5
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where Yrepresents the dependent variable, Xthe independent variable, and
ϵthe error term. After conducting a hypothesis test, we obtain the following
results: the test statistic is t= 2.76 and the critical value at the 5
Given this information, can we reject the null hypothesis at a significance
level of 5
Solution
Step 1: State the null and alternative hypotheses.
The null hypothesis states that the slope coefficient is equal to 0 (H0:β1= 0),
while the alternative hypothesis states that the slope coefficient is not equal to
0 (Ha:β1= 0).
Step 2: Determine the decision rule.
Since the test statistic t= 2.76, and the critical value at the 5
Step 3: Make a decision.
Since |2.76|>2.262, we reject the null hypothesis.
Step 4: State the conclusion.
Therefore, at the 5
Question 6
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where the error term ϵfollows a normal distribution with mean 0 and variance
4
σ2. In a study examining the relationship between hours studied (indepen-
dent variable) and exam score (dependent variable), a researcher obtained the
following results for the regression analysis:
ˆ
β0= 60.5,ˆ
β1= 3.2,
s2= 9.6,
n= 30.
Determine a 95% confidence interval for the true slope β1.
Solution
Step 1: The formula to compute the confidence interval for the slope estimate
β1is given by:
ˆ
β1±tα/2,n−2·s
√∑n
i=1(Xi−¯
X)2
,
where tα/2,n−2is the critical value from the t-distribution with n−2degrees of
freedom.
Step 2: First, we compute the critical value tα/2,n−2at the 95% confidence
level. Since α= 0.05, the degrees of freedom are n−2 = 30 −2 = 28. Using a
t-table or a statistical software, we find t0.025,28 ≈2.048.
Step 3: Next, we calculate the standard error of the estimate:
SE(ˆ
β1) = s
√∑n
i=1(Xi−¯
X)2
.
Step 4: Given s2= 9.6, we have s=√9.6 = 3.1. Since we do not have
the individual data values, we cannot directly compute the sum of squares of
(Xi−¯
X). However, we can use the fact that ∑n
i=1(Xi−¯
X)2= (n−1) ·s2
X,
where s2
Xis the sample variance of the independent variable.
Step 5: We can estimate s2
Xusing s2, which gives s2
X=s2
n−1=9.6
30−1≈0.32.
Therefore, the standard error of the estimate is SE(ˆ
β1) = 3.1
√0.32 ≈1.75.
Step 6: Finally, we can calculate the confidence interval by plugging into the
formula:
3.2±2.048 ×1.75.
Step 7: Thus, the 95% confidence interval for the true slope parameter β1is
approximately [0.595,5.805].
Question 7
Question
Let Ybe a random variable representing the response variable and Xbe a
random variable representing the predictor variable in a linear regression model.
5
Suppose we have the following simple linear regression model:
Y=β0+β1X+ε,
where εis the error term with mean 0 and variance σ2. Given a sample of size
n= 20, the following data was obtained:
X Y
1 2.3
2 4.5
3 6.7
4 8.9
5 11.1
6 13.3
7 15.5
8 17.7
9 19.9
10 22.1
11 24.3
12 26.5
13 28.7
14 30.9
15 33.1
16 35.3
17 37.5
18 39.7
19 41.9
20 44.1
Determine the least squares estimators ˆ
β0and ˆ
β1, the standard errors of
these estimators, and the 95
Solution
Step 1: Compute the least squares estimators ˆ
β0and ˆ
β1using the formulas:
ˆ
β1=∑n
i=1(Xi−¯
X)(Yi−¯
Y)
∑n
i=1(Xi−¯
X)2,
ˆ
β0=¯
Y−ˆ
β1¯
X,
where ¯
Xand ¯
Yare the sample means of Xand Yrespectively.
Step 2: Calculate the sample means:
¯
X=1
n
n
∑
i=1
Xi= 10.5,
6
¯
Y=1
n
n
∑
i=1
Yi= 22.2.
Step 3: Compute the least squares estimators:
ˆ
β1=∑20
i=1(Xi−10.5)(Yi−22.2)
∑20
i=1(Xi−10.5)2=203.4
215 ≈0.94651,
ˆ
β0= 22.2−0.94651 ×10.5≈11.49355.
Step 4: Compute the residuals ˆεi=Yi−ˆ
β0−ˆ
β1Xifor each observation.
Step 5: Calculate the standard errors of ˆ
β0and ˆ
β1using the formulas:
SE(ˆ
β0) = ˆσ√1
n+¯
X2
∑n
i=1(Xi−¯
X)2,
SE(ˆ
β1) = ˆσ
√∑n
i=1(Xi−¯
X)2
,
where ˆσis the estimate of the standard deviation of the error term.
Step 6: Finally, compute the 95
ˆ
β0−tα/2,n−2×SE(ˆ
β0)≤β0≤ˆ
β0+tα/2,n−2×SE(ˆ
β0),
ˆ
β1−tα/2,n−2×SE
Question 8
Question
Let Yidenote the response variable and Xibe the predictor variable for i=
1,2, . . . , n observations. Consider the simple linear regression model given by
Yi=β0+β1Xi+εiwhere εiare independent and identically distributed with
mean zero and variance σ2. Suppose we have fitted the model and obtained the
following output:
ˆ
Y= 4.2+2.6X, se= 1.4, n = 20
(a) Test at the 5% significance level whether the slope coefficient is signifi-
cant.
(b) Construct a 95% confidence interval for the slope coefficient.
7
Solution
(a) To test whether the slope coefficient is significant, we need to perform a
hypothesis test. The hypotheses are:
H0:β1= 0 (slope coefficient is not significant)
Ha:β1= 0 (slope coefficient is significant)
We will use the t-test statistic given by:
t=ˆ
β1
se/√∑(Xi−¯
X)2=2.6
1.4/√∑(Xi−¯
X)2
Under H0,tfollows a t-distribution with n−2 = 18 degrees of freedom. The
critical value at 5% significance level for a two-tailed test is tcritical =±2.101.
If |t|> tcritical, we reject H0; otherwise, we fail to reject H0. Let’s calculate
tand make a decision.
Step 1: Calculate the sum of squares of (Xi−¯
X):
∑(Xi−¯
X)2=∑X2
i−1
n(∑Xi)2
Step 2: Substitute the values into the formula for t:
t=2.6
1.4/√∑(Xi−¯
X)2
Step 3: Compare |t|with tcritical to make a decision.
(b) To construct a 95% confidence interval for the slope coefficient β1, we
use the formula: ˆ
β1±tα/2,n−2×se
√∑(Xi−¯
X)2
Where tα/2,n−2is the critical t-value for a 95% confidence interval with
n−2degrees of freedom. The confidence interval is calculated by substituting
the values into the formula.
Step 1: Calculate the critical t-value tα/2,n−2.
Step 2: Substitute the values into the formula to find the confidence interval
for β1.
Question 9
Question
Suppose we have collected data on a sample of 50 individuals, including their
annual income and the number of years of education they have completed. The
least squares regression line for predicting annual income based on years of
education is given by ˆ
Y= 2000 + 1000X, where Yrepresents annual income
8
(in dollars) and Xrepresents years of education. The standard error of the
estimate for this regression line is 500 dollars. Test the null hypothesis that the
true slope coefficient is equal to 0 against the alternative hypothesis that the
true slope coefficient is not equal to 0, at a significance level of 0.05.
Solution
Step 1: State the hypotheses: - Null Hypothesis (H0): The true slope coefficient
is equal to 0 (β1= 0). - Alternative Hypothesis (H1): The true slope coefficient
is not equal to 0 (β1= 0).
Step 2: Compute the test statistic: The test statistic for testing the slope
coefficient in linear regression is given by:
t=b1
SE(b1)
where b1is the estimate of the slope coefficient and SE(b1)is the standard error
of the estimate.
In this case, b1= 1000 and SE(b1) = 500. Thus, the test statistic is:
t=1000
500 = 2
Step 3: Determine the degrees of freedom: The degrees of freedom for testing
the slope coefficient in linear regression is equal to n−2, where nis the number
of observations in the sample. Here, n= 50, so the degrees of freedom is
50 −2 = 48.
Step 4: Determine the critical value: Since we are testing at a significance
level of 0.05, the critical values for a two-tailed test with 48 degrees of freedom
are approximately ±2.0106.
Step 5: Make a decision: Since the calculated test statistic of 2 does not fall
in the rejection region defined by the critical values of ±2.0106, we do not reject
the null hypothesis.
Step 6: Conclusion: We do not have enough evidence to conclude that the
true slope coefficient is significantly different from 0 at a significance level of
0.05.
Question 10
Question
Suppose we have a dataset consisting of 20 observations and 3 predictor vari-
ables. The fitted linear regression model is given by ˆ
Y= 2X1−3X2+4X3. The
residual standard error is found to be 5. Perform a hypothesis test to determine
if the coefficient of X2is significantly different from zero at the 5
9
Solution
Step 1: State the null and alternative hypotheses.
Let β2be the coefficient of X2. The hypotheses are: - Null hypothesis:
H0:β2= 0 - Alternative hypothesis: H1:β2= 0
Step 2: Calculate the t-statistic. The t-statistic is given by:
t=ˆ
β2
SE(ˆ
β2)
where ˆ
β2is the estimated coefficient of X2and SE(ˆ
β2)is the standard error
of the coefficient estimate.
Step 3: Find the p-value associated with the t-statistic. Under the null
hypothesis, the t-statistic follows a t-distribution with n−p−1 = 20−3−1 = 16
degrees of freedom (where nis the number of observations and pis the number
of predictor variables). We can find the p-value by comparing the absolute value
of the t-statistic to the critical t-value for a two-tailed test.
Step 4: Make a decision based on the p-value. At the 5
Step 5: Interpret the results. If the null hypothesis is rejected, we can
conclude that the coefficient of X2is significantly different from zero.
Question 11
Question
Let Ybe a response variable and Xbe a predictor variable in a simple linear
regression model. The data below was collected from a random sample of size
n= 10:X Y
2 5
4 7
6 10
8 14
10 13
12 16
14 19
16 20
18 21
20 25
Calculate a 95% confidence interval for the slope of the regression line.
Solution
Step 1: Calculate the sample mean of X(X), the sample mean of Y(Y), the
sample variance of X(s2
X), the sample variance of Y(s2
Y), and the sample
10
covariance between Xand Y(sXY ). Using the formulas:
X=1
n
n
∑
i=1
Xi, Y =1
n
n
∑
i=1
Yi
s2
X=1
n−1
n
∑
i=1
(Xi−X)2, s2
Y=1
n−1
n
∑
i=1
(Yi−Y)2
sXY =1
n−1
n
∑
i=1
(Xi−X)(Yi−Y)
we calculate:
X= 11, Y = 15.5
s2
X= 44.44, s2
Y= 35.33
sXY = 42.5
Step 2: Calculate the slope estimator, b1, using the formula:
b1=sXY
s2
X
Substitute the values into the formula:
b1=42.5
44.44 = 0.958
Step 3: Calculate the standard error of the slope (SEb1) using the formula:
SEb1=√s2
Y
n−2/√s2
X
Substitute the values into the formula:
SEb1=√35.33
8/√44.44 = 0.680
Step 4: Find the critical value t∗for a 95% confidence interval with n−2 = 8
degrees of freedom. From the t-distribution table, t∗= 2.306.
Step 5: Calculate the margin of error at a 95% confidence level:
ME =t∗×SEb1= 2.306 ×0.680 = 1.568
Step 6: Finally, construct the 95% confidence interval for the slope:
CI = (b1−M E, b1+ME) = (0.958 −1.568,0.958 + 1.568) = (−0.610,2.526)
Therefore, the 95% confidence interval for the slope of the regression line is
(−0.610,2.526).
11
Question 12
Question
Suppose we have a linear regression model given by Y=β0+β1X+ε, where
Yis the response variable, Xis the predictor variable, β0and β1are unknown
coefficients, and εis the error term. Given the following information from a
sample of size n= 20:
20
∑
i=1
Yi= 360,
20
∑
i=1
Xi= 30,
20
∑
i=1
Y2
i= 2440,
20
∑
i=1
X2
i= 118,
20
∑
i=1
XiYi= 450
Determine the estimated regression line of Yon Xand conduct a hypothesis
test to determine if there is a significant linear relationship between Xand Yat
the 95% confidence level. Assume all necessary assumptions for linear regression
are satisfied.
Solution
Step 1: Calculate the slope ( ˆ
β1) and intercept ( ˆ
β0) of the estimated regression
line using the formula:
ˆ
β1=n∑n
i=1 XiYi−∑n
i=1 Xi∑n
i=1 Yi
n∑n
i=1 X2
i−(∑n
i=1 Xi)2
ˆ
β0=∑n
i=1 Yi−ˆ
β1∑n
i=1 Xi
n
Substitute the given values into the formulas:
ˆ
β1=20(450) −30(360)
20(118) −302=9000 −10800
2360 −900 =−1800
1460 ≈ −1.23
ˆ
β0=360 −(−1.23)(30)
20 ≈55.6
Therefore, the estimated regression line is ˆ
Y= 55.6−1.23X.
Step 2: Perform a hypothesis test to determine if there is a significant linear
relationship between Xand Yat the 95% confidence level.
The hypotheses are: - H0:β1= 0 (There is no linear relationship between
Xand Y) - H1:β1= 0 (There is a linear relationship between Xand Y)
We will conduct a t-test using the test statistic:
t=ˆ
β1
√MSE
∑n
i=1(Xi−X)2
=−1.23
√2440−(−1.23)(450)
20−2·1
20(118)−302≈ −2.46
12
Under the null hypothesis, tfollows a t-distribution with n−2 = 18 degrees
of freedom. At the 95% confidence level, the critical value for a two-tailed test
is approximately ±2.101.
Since |t|= 2.46 >2.101, we reject the null hypothesis. Therefore, we have
enough evidence to conclude that there is a significant linear relationship be-
tween Xand Yat the 95% confidence level.
Question 14
Question
Let Yibe the response variable and Xibe the predictor variable. Consider the
following simple linear regression model:
Yi=β0+β1Xi+εi
where εi∼N(0, σ2)are independent and identically distributed. From a
sample of size n= 20, the following least squares regression line was obtained:
ˆ
Y= 5.1+2.3X. Further analysis produced the following ANOVA table:
Source Sum of Squares Degrees of Freedom Mean Square F
Regression 78.56 1 78.56 24.9
Residual 41.28 18 2.29
Total 119.84 19
Perform a hypothesis test at significance level α= 0.05 to determine whether
there is a significant linear relationship between the response variable and the
predictor variable.
Solution
Step 1: State the Hypotheses
The null hypothesis H0and the alternative hypothesis HAare: - H0:β1= 0
(There is no linear relationship between the response variable and the predictor
variable) - HA:β1= 0 (There is a linear relationship between the response
variable and the predictor variable)
Step 2: Calculate the F-statistic
The F-statistic is calculated as Mean Square Regression
Mean Square Residual =78.56
2.29 ≈34.25.
Step 3: Find the Critical Value
With degrees of freedom for the regression and residual being 1 and 18
respectively, and significance level α= 0.05, the critical value can be found using
an F-distribution table. For α= 0.05, the critical values are approximately 6.39
and 0.156.
Step 4: Make a Decision
Since the F-statistic (34.25) is greater than the critical value (6.39), we reject
the null hypothesis.
13
Step 5: State the Conclusion
There is enough evidence to conclude that there is a significant linear rela-
tionship between the response variable and the predictor variable at a signifi-
cance level of 0.05.
Question 15
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+εi, where
εiare independent identically distributed random variables with mean 0and
variance σ2.
Given the following data:
X Y
1 2
2 4
3 6
4 8
5 10
Estimate the regression coefficients β0and β1using the method of least
squares.
Solution
Step 1: We first calculate the sample means ¯
Xand ¯
Yusing the given data.
¯
X=1+2+3+4+5
5= 3 and ¯
Y=2+4+6+8+10
5= 6
Step 2: Next, we calculate the sample covariance sXY and the sample vari-
ance s2
X.
sXY =∑n
i=1(Xi−¯
X)(Yi−¯
Y)
n−1=(1 −3)(2 −6) + (2 −3)(4 −6) + (3 −3)(6 −6) + (4 −3)(8 −6) + (5 −3)(10 −6)
5−1= 4
s2
X=∑n
i=1(Xi−¯
X)2
n−1=(1 −3)2+ (2 −3)2+ (3 −3)2+ (4 −3)2+ (5 −3)2
5−1= 2.5
Step 3: Calculate the estimated regression coefficient ˆ
β1.
ˆ
β1=sXY
s2
X
=4
2.5= 1.6
Step 4: Use the estimated coefficient ˆ
β1to find the estimated intercept ˆ
β0.
ˆ
β0=¯
Y−ˆ
β1¯
X= 6 −1.6(3) = 1.8
Therefore, the estimated regression equation is ˆ
Y= 1.8+1.6X.
14
Question 16
Question
Let Yrepresent the dependent variable and Xrepresent the independent vari-
able in a simple linear regression model. A sample of 20 data points yielded the
following results:
∑Y= 400,∑X= 50,∑Y2= 16000,∑X2= 600,and ∑XY = 2500
Using these results, test at the 5
Solution
Step 1: The null and alternative hypotheses are stated as:
H0:β= 0 (slope is equal to 0)
HA:β= 0 (slope is not equal to 0)
Step 2: Calculate the sample size, the coefficient of determination R2, and
the test statistic:
n= 20
R2=∑n
i=1(ˆ
Yi−¯
Y)2
∑n
i=1(Yi−¯
Y)2=∑Y2−(∑Y)2
n
∑Y2= 1−(400
20 )2
16000
20
= 1−1002
800 = 1−10000
800 = 1−12.5 = −11.5
t=ˆ
β−0
se(ˆ
β)=
∑XY −∑X∑Y
n
∑X2−(∑X)2
n
√∑Y2−(∑Y)2
n−ˆ
β2(∑X2−(∑X)2
n)
n−2
=
2500−50·400
20
600−502
20
√16000−4002
20 −ˆ
β2(600−502
20 )
20−2
=
=2500 −2000
20
600 −2500
20
∇·√16000 −160000
20 −ˆ
β2(600 −2500
20 )
18 =2500 −100
600 −125 ∇·√16000 −8000 −ˆ
β2(600 −125)
18 =2400
475 ∇·√8000 −600 ˆ
β2
18
Step 3: Reject the null hypothesis if |t|> tn−2,α/2. Since this is a two-tailed
test at the 5
In this case, we calculate the value of |t|and compare it to the critical value.
Step 4: Calculate the p-value associated with the test statistic:
p-value = 2(1−P(tn−2≤ |t|)) = 2(1−P(t18 ≤2.097)) = 2(1−0.9756) = 2(0.0244) = 0.0488
Step 5: Make a decision: Since p-value = 0.0488 <0.05 and |t|= 2.097 >
2.101, we reject the null hypothesis.
Step 6: Conclusion: There is sufficient evidence at the 5
15
Question 17
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εi∼N(0, σ2)for i= 1,2, . . . , n. The least squares estimates for the
model coefficients are ˆ
β0= 2 and ˆ
β1= 3.5. Given that the variance of the error
terms is σ2= 4, test the hypothesis H0:β1= 3 vs Ha:β1= 3 at a significance
level of α= 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing the slope
coefficient in simple linear regression is given by:
t=ˆ
β1−β1,null
√Var(ˆ
β1)
Where β1,null is the hypothesized value of the slope coefficient under the
null hypothesis (in this case, β1,null = 3). The variance of the estimated slope
coefficient is given by:
Var(ˆ
β1) = σ2
∑n
i=1(Xi−¯
X)2
Given that σ2= 4, we need to calculate ∑n
i=1(Xi−¯
X)2. Without further
information, we are unable to calculate this directly. Therefore, we cannot
calculate the exact test statistic.
Step 2: Determine the critical value. Since we cannot calculate the exact test
statistic, we are unable to determine the critical value for the t-distribution with
n−2degrees of freedom. Hence, we are unable to proceed with the hypothesis
test without the additional information required for the calculations.
Therefore, the hypothesis test cannot be conducted without the necessary
information to calculate the test statistic and critical value accurately.
Question 18
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εiare independently and identically distributed with mean 0 and variance
σ2. A random sample of 10 observations produced the following estimated
regression equation: ˆ
Y= 12.5 + 1.6Xand a coefficient of determination R2=
0.82. Test the hypothesis H0:β1= 0 against Ha:β1= 0 at the 5% level of
significance.
16
Solution
Step 1: First, we need to find the test statistic. The test statistic is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of β1and SE(ˆ
β1)is the standard error of the estimate.
Step 2: We are given that ˆ
Y= 12.5+1.6X. From this, we can see that
ˆ
β1= 1.6.
Step 3: To find SE(ˆ
β1), we use the formula:
SE(ˆ
β1) = √MSE
SXX
where MSE is the mean squared error and SXX is the sum of squares Xi−¯
X.
Step 4: Given that R2= 0.82, we can find M SE using the formula:
MSE =(n−2) ×SSE
n−2=(10 −2) ×18.4
10 −2= 16.56
where nis the number of observations and SSE is the sum of squared errors.
Step 5: We can find SXX using the formula:
SXX =∑(Xi−¯
X)2= 122.5
where ¯
Xis the mean of the Xivalues.
Step 6: Substituting the values into the formula, we get:
SE(ˆ
β1) = √16.56
122.5≈0.3888
Step 7: Now, we can find the test statistic:
t=1.6
0.3888 ≈4.1171
Step 8: Since this is a two-tailed test, we will compare this value with the
critical value at the 5
Step 9: The critical values are approximately ±2.306.
Step 10: Since |4.1171|>2.306, we reject the null hypothesis H0:β1= 0.
Therefore, we have sufficient evidence to conclude that β1is not equal to 0 at
the 5% significance level.
Question 19
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ε,
where ε∼N(0, σ2).
For a specific dataset, we obtain the following results:
17
Coefficient Estimate Standard Error
β04.2 0.6
β12.8 0.4
Perform a hypothesis test to determine if there is a significant linear rela-
tionship between Xand Yat a 5
Solution
Step 1: Formulate the hypotheses
The null hypothesis is that there is no linear relationship between Xand Y:
H0:β1= 0
The alternative hypothesis is that there is a significant linear relationship be-
tween Xand Y:
Ha:β1= 0
Step 2: Calculate the test statistic
The test statistic for testing the slope coefficient β1is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of coefficient β1and SE(ˆ
β1)is the standard error of
the estimate.
In this case,
t=2.8
0.4= 7
Step 3: Determine the critical value
Since the alternative hypothesis is two-sided, we will look for the critical values
in the t-distribution table for a two-tailed test with α= 0.05 and degrees of
freedom n−2(for simple linear regression).
For a significance level of 5
Step 4: Make a decision
Since the test statistic t = 7 is greater than 2.306, we reject the null hypothesis.
Therefore, we have enough evidence to conclude that there is a significant
linear relationship between Xand Y.
Question 20
Question
Suppose a linear regression model is fitted to a set of data and the estimated
regression equation is ˆy= 3.2 + 2.5x. The standard error of the estimate is
4.8. Test the null hypothesis H0:β1= 0 against the alternative hypothesis
Ha:β1= 0 at the 5
18
Solution
Step 1: The test statistic tis calculated using the formula:
t=b1
SE(b1)
where b1is the estimated coefficient for variable x, and SE(b1)is the standard
error of this coefficient.
Step 2: Given that the estimated coefficient b1= 2.5and the standard error
SE(b1) = 4.8, we can substitute these values into the formula to find the test
statistic:
t=2.5
4.8
Step 3: Calculate the value of t:
t≈0.52
Step 4: Degrees of freedom for the test statistic tis equal to the sample size
minus 2, where 2 is the number of parameters estimated in the regression model
(intercept and slope).
df =n−2
Step 5: Since the significance level is 5
Step 6: Compare the calculated tvalue with the critical values. Since 0.52
falls within the range (-2.571, 2.571), we fail to reject the null hypothesis H0:
β1= 0.
Step 7: Therefore, at 5
Question 21
Question
Let Yidenote the response variable and Xidenote the predictor variable for a
linear regression model. Consider the following multiple linear regression model:
Yi=β0+β1X1i+β2X2i+εi
where εi∼N(0, σ2)are independent and identically distributed.
Given the data set (X1i, X2i, Yi)for i= 1,2,3, . . . , n, we are interested in
testing the null hypothesis H0:β1=β2against the alternative hypothesis
Ha:β1=β2using a significance level of α= 0.05. The fitted model results in
ˆ
β1= 3.9,ˆ
β2= 4.1, and the standard error of ˆ
β1is 0.6 while the standard error
of ˆ
β2is 0.5.
Perform a hypothesis test for β1=β2using the given information.
19
Solution
Step 1: Formulate the test statistic. Given that we have to test whether β1=β2,
the test statistic is computed as
t=ˆ
β1−ˆ
β2
√SE(ˆ
β1)2+SE(ˆ
β2)2
Step 2: Compute the test statistic. Plugging in the given values, we obtain
t=3.9−4.1
√0.62+ 0.52
Step 3: Determine the critical value. Since we are performing a two-tailed
test at a significance level of α= 0.05, the critical values are ±zα/2where
zα/2=z0.025 ≈1.96.
Step 4: Make a decision. If |t|>1.96, we reject the null hypothesis H0:
β1=β2. Otherwise, we fail to reject H0.
Step 5: Calculate the test statistic value. Now, substitute the values and
compute tto determine the decision:
t=3.9−4.1
√0.62+ 0.52=−0.2
√0.36 + 0.25 ≈−0.2
√0.61 ≈ −0.255
Since | − 0.255|<1.96, we fail to reject H0.
Step 6: Formulate the conclusion. Since the test statistic value falls within
the range of the critical values, we do not have enough evidence to conclude
that β1is different from β2.
Question 22
Question
Let Y1, Y2, . . . , Ynbe independent and identically distributed random variables
following a normal distribution with mean µand variance σ2. Consider the
simple linear regression model Yi=β0+β1Xi+εi, where β0,β1, and σ2are
unknown parameters to be estimated.
Suppose that the least squares estimates of β0and β1are denoted as ˆ
β0and
ˆ
β1, respectively. Show that ˆ
β0and ˆ
β1are unbiased estimators for β0and β1,
respectively.
Solution
Step 1: Recall that the least squares estimates of β0and β1are given by:
ˆ
β1=
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
20
ˆ
β0=¯
Y−ˆ
β1¯
X
Step 2: Let’s start by showing that ˆ
β1is an unbiased estimator for β1. The
expectation of ˆ
β1can be computed as follows:
E(ˆ
β1) = E
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)E(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−E(¯
Y))
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−µ)
n
∑
i=1
(Xi−¯
X)2
Step 3: Since the Yi’s are independent and identically distributed with mean
µ, we have E(Yi) = µfor all i. Thus, E(ˆ
β1) =
n
∑
i=1
(Xi−¯
X)·0
n
∑
i=1
(Xi−¯
X)2= 0.
Step 4: Therefore, ˆ
β1is an unbiased estimator for β1.
Step 5: Next, we will show that ˆ
β0is an unbiased estimator for β0. By
linearity of expectation, we have:
E(ˆ
β0) = E(¯
Y)−E(ˆ
β1¯
X)
Step 6: Since ¯
Yand ¯
Xare constants, we can simplify this expression to:
E(ˆ
β0) = ¯
Y−¯
XE(ˆ
β1)
Step 7: Since ˆ
β1is an unbiased estimator for β1, we have E(ˆ
β1) = 0. Thus,
E(ˆ
β0) = ¯
Y−0·¯
X=¯
Y.
Step 8: Therefore, ˆ
β0is an unbiased estimator for β0.
Question 23
Question
Suppose you are given the following multiple regression model:
y= 3x1−2x2+ 5x3+ 4
21
where yis the dependent variable, and x1,x2, and x3are the independent
variables. You are also given the following outputs from the regression analysis:
• Residual standard error = 2.5
•R2= 0.85
• F-statistic = 15.6
Is this model statistically significant at the α= 0.05 level? Justify your answer.
Solution
Step 1: Hypotheses
The null hypothesis H0is that all the coefficients in the regression model are
equal to zero:
H0:β1=β2=β3= 0
The alternative hypothesis H1is that at least one coefficient is not equal to
zero:
H1: At least one of β1,β2,β3is not equal to zero.
Step 2: F-statistic test
First, we calculate the degrees of freedom for the numerator and denominator
of the F-statistic:
• Numerator degrees of freedom = number of coefficients being tested = 3
• Denominator degrees of freedom = number of observations - number of
coefficients = n−k−1, where nis the number of observations and kis
the number of coefficients including the intercept term. Since we do not
have the sample size, we cannot perform the hypothesis test.
Step 3: Decision
We cannot perform the F-test without knowing the sample size n. Therefore, we
are unable to determine the statistical significance of the model at the α= 0.05
level without additional information.
Question 24
Question
Let’s consider a multiple linear regression model with three predictors X1,X2,
and X3. The estimated regression equation is given by:
ˆ
Y= 15 + 3X1−2X2+ 4X3
Assume the estimated variances for the coefficient estimates are V ar(ˆ
β0) = 10,
V ar(ˆ
β1) = 4,V ar(ˆ
β2) = 6, and V ar(ˆ
β3) = 8. Suppose we want to test the
following null hypothesis:
H0:β1= 0, H1:β1= 0
22
Perform a hypothesis test to determine whether there is a linear relationship
between X1and Yusing a significance level of 0.05.
Solution
Step 1: The test statistic for testing the hypothesis βj= 0 versus Ha:βj= 0
is given by:
t=ˆ
βj
√V ar(ˆ
βj)
Step 2: In this case, we are interested in testing the hypothesis H0:β1= 0
versus H1:β1= 0. The test statistic for β1is:
t=3
√4=3
2
Step 3: The critical value for a two-tailed test at a 5
Step 4: Using a statistical table, we find that t0.025,16 ≈2.12.
Step 5: Compare the absolute value of the test statistic to the critical value.
Since 3
2= 1.5<2.12, we do not reject the null hypothesis.
Step 6: Therefore, we do not have enough evidence to conclude that there
is a linear relationship between X1and Yat a significance level of 0.05.
Question 25
Question
Let Ybe the dependent variable and Xbe the independent variable in a
linear regression model. Suppose that the estimated regression equation is
ˆ
Y= 3 + 1.5X. The standard error of the estimate is found to be 2.5. Given
this information, test the null hypothesis H0:β1= 0 against the alternative
hypothesis Ha:β1= 0 at the 5
Solution
Step 1: Determine the test statistic using the formula:
t=ˆ
β1−0
SE(ˆ
β1)
where ˆ
β1= 1.5and SE(ˆ
β1) = 2.5. Plugging in these values, we get:
t=1.5−0
2.5= 0.6
Step 2: Find the degrees of freedom. Since we are testing the slope coeffi-
cient, the degrees of freedom is n−2, where nis the sample size. However, the
sample size is not given in the question.
23
Step 4: Calculate the test statistic value. We have SSR =β2
1×SSXX and
SSE =SSE/df where SSXX is the sum of squares of Xand df is the degrees of
freedom for the error.
Plugging in the given values, we find:
SSR = 1.52×SSXX
SSE = 4.0/18
Step 5: Determine the critical value. Look up the critical value from the
F-distribution table with degrees of freedom 1 and 18 at the 5% significance
level. Let’s say the critical value is Fcrit.
Step 6: Make a decision. If F > Fcrit, reject the null hypothesis. Otherwise,
fail to reject the null hypothesis.
Step 7: Conclusion. State your conclusion based on the decision made in
step 6.
Question 2
Question
Suppose we have a dataset with the following linear regression model:
Y= 3X1−2X2+ 5X3+ϵ
Where X1, X2, and X3are independent variables, and ϵis the error term. Af-
ter performing linear regression analysis, we obtained the following estimated
regression coefficients:
ˆ
β1= 2.6,ˆ
β2=−1.8,ˆ
β3= 4.9
Also, the standard errors of these estimates were found to be:
SE(ˆ
β1) = 0.3, SE(ˆ
β2) = 0.4, SE(ˆ
β3) = 0.2
Using a significance level of 0.05, test the null hypothesis that the coefficient for
X2is equal to zero.
Solution
Step 1: Formulate the hypotheses.
The null hypothesis (H0) and alternative hypothesis (H1) can be formulated as:
H0:β2= 0
H1:β2= 0
2
Step 2: Calculate the test statistic.
The test statistic for testing the significance of β2is given by:
t=ˆ
β2
SE(ˆ
β2)
Substitute the given values to find the test statistic:
t=−1.8
0.4=−4.5
Step 3: Determine the degrees of freedom.
The degrees of freedom for the t-distribution in this case is n−3, where nis the
number of observations. Since the degrees of freedom is not explicitly given in
the question, we assume that there are enough observations to ensure that the
t-distribution is a good approximation.
Step 4: Determine the critical value.
Using a significance level of 0.05 and the two-tailed test, the critical value is
found using a t-table with appropriate degrees of freedom.
Step 5: Make a decision.
Compare the absolute value of the test statistic with the critical value from the
t-table. If the absolute value of the test statistic is greater than the critical value,
we reject the null hypothesis; otherwise, we fail to reject the null hypothesis.
Step 6: Conclusion.
Based on the decision in Step 5, make a conclusion regarding the null hypothesis
at the 0.05 significance level.
Question 4
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+ϵi, where
ϵi∼N(0, σ2)represents the error term. We are interested in testing the null
hypothesis H0:β1= 0 against the alternative hypothesis H1:β1= 0. Given
the following data and output summary from a regression analysis:
Variable ˆ
βSE(ˆ
β)
β03.56 0.42
β10.78 0.06
Perform the hypothesis test using a significance level of 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing whether the
slope is significantly different from zero is given by:
t=ˆ
β1−0
SE(ˆ
β1)
3
Given ˆ
β1= 0.78 and SE(ˆ
β1) = 0.06, we have:
t=0.78 −0
0.06 = 13
Step 2: Determine the critical value. Since we are testing for a two-tailed
hypothesis test at a significance level of 0.05, the critical value can be obtained
using a t-distribution with n−2degrees of freedom.
For a two-tailed test with α= 0.05 and n−2 = 10 −2=8degrees of
freedom, the critical values are ±2.306.
Step 3: Make a decision. Since |t|= 13 >2.306, we reject the null hypothesis
H0:β1= 0 in favor of the alternative hypothesis H1:β1= 0 at the 0.05
significance level. This indicates that the slope coefficient β1is significantly
different from zero.
Question 5
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where Yrepresents the dependent variable, Xthe independent variable, and
ϵthe error term. After conducting a hypothesis test, we obtain the following
results: the test statistic is t= 2.76 and the critical value at the 5
Given this information, can we reject the null hypothesis at a significance
level of 5
Solution
Step 1: State the null and alternative hypotheses.
The null hypothesis states that the slope coefficient is equal to 0 (H0:β1= 0),
while the alternative hypothesis states that the slope coefficient is not equal to
0 (Ha:β1= 0).
Step 2: Determine the decision rule.
Since the test statistic t= 2.76, and the critical value at the 5
Step 3: Make a decision.
Since |2.76|>2.262, we reject the null hypothesis.
Step 4: State the conclusion.
Therefore, at the 5
Question 6
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where the error term ϵfollows a normal distribution with mean 0 and variance
4
σ2. In a study examining the relationship between hours studied (indepen-
dent variable) and exam score (dependent variable), a researcher obtained the
following results for the regression analysis:
ˆ
β0= 60.5,ˆ
β1= 3.2,
s2= 9.6,
n= 30.
Determine a 95% confidence interval for the true slope β1.
Solution
Step 1: The formula to compute the confidence interval for the slope estimate
β1is given by:
ˆ
β1±tα/2,n−2·s
√∑n
i=1(Xi−¯
X)2
,
where tα/2,n−2is the critical value from the t-distribution with n−2degrees of
freedom.
Step 2: First, we compute the critical value tα/2,n−2at the 95% confidence
level. Since α= 0.05, the degrees of freedom are n−2 = 30 −2 = 28. Using a
t-table or a statistical software, we find t0.025,28 ≈2.048.
Step 3: Next, we calculate the standard error of the estimate:
SE(ˆ
β1) = s
√∑n
i=1(Xi−¯
X)2
.
Step 4: Given s2= 9.6, we have s=√9.6 = 3.1. Since we do not have
the individual data values, we cannot directly compute the sum of squares of
(Xi−¯
X). However, we can use the fact that ∑n
i=1(Xi−¯
X)2= (n−1) ·s2
X,
where s2
Xis the sample variance of the independent variable.
Step 5: We can estimate s2
Xusing s2, which gives s2
X=s2
n−1=9.6
30−1≈0.32.
Therefore, the standard error of the estimate is SE(ˆ
β1) = 3.1
√0.32 ≈1.75.
Step 6: Finally, we can calculate the confidence interval by plugging into the
formula:
3.2±2.048 ×1.75.
Step 7: Thus, the 95% confidence interval for the true slope parameter β1is
approximately [0.595,5.805].
Question 7
Question
Let Ybe a random variable representing the response variable and Xbe a
random variable representing the predictor variable in a linear regression model.
5
Suppose we have the following simple linear regression model:
Y=β0+β1X+ε,
where εis the error term with mean 0 and variance σ2. Given a sample of size
n= 20, the following data was obtained:
X Y
1 2.3
2 4.5
3 6.7
4 8.9
5 11.1
6 13.3
7 15.5
8 17.7
9 19.9
10 22.1
11 24.3
12 26.5
13 28.7
14 30.9
15 33.1
16 35.3
17 37.5
18 39.7
19 41.9
20 44.1
Determine the least squares estimators ˆ
β0and ˆ
β1, the standard errors of
these estimators, and the 95
Solution
Step 1: Compute the least squares estimators ˆ
β0and ˆ
β1using the formulas:
ˆ
β1=∑n
i=1(Xi−¯
X)(Yi−¯
Y)
∑n
i=1(Xi−¯
X)2,
ˆ
β0=¯
Y−ˆ
β1¯
X,
where ¯
Xand ¯
Yare the sample means of Xand Yrespectively.
Step 2: Calculate the sample means:
¯
X=1
n
n
∑
i=1
Xi= 10.5,
6
¯
Y=1
n
n
∑
i=1
Yi= 22.2.
Step 3: Compute the least squares estimators:
ˆ
β1=∑20
i=1(Xi−10.5)(Yi−22.2)
∑20
i=1(Xi−10.5)2=203.4
215 ≈0.94651,
ˆ
β0= 22.2−0.94651 ×10.5≈11.49355.
Step 4: Compute the residuals ˆεi=Yi−ˆ
β0−ˆ
β1Xifor each observation.
Step 5: Calculate the standard errors of ˆ
β0and ˆ
β1using the formulas:
SE(ˆ
β0) = ˆσ√1
n+¯
X2
∑n
i=1(Xi−¯
X)2,
SE(ˆ
β1) = ˆσ
√∑n
i=1(Xi−¯
X)2
,
where ˆσis the estimate of the standard deviation of the error term.
Step 6: Finally, compute the 95
ˆ
β0−tα/2,n−2×SE(ˆ
β0)≤β0≤ˆ
β0+tα/2,n−2×SE(ˆ
β0),
ˆ
β1−tα/2,n−2×SE
Question 8
Question
Let Yidenote the response variable and Xibe the predictor variable for i=
1,2, . . . , n observations. Consider the simple linear regression model given by
Yi=β0+β1Xi+εiwhere εiare independent and identically distributed with
mean zero and variance σ2. Suppose we have fitted the model and obtained the
following output:
ˆ
Y= 4.2+2.6X, se= 1.4, n = 20
(a) Test at the 5% significance level whether the slope coefficient is signifi-
cant.
(b) Construct a 95% confidence interval for the slope coefficient.
7
Solution
(a) To test whether the slope coefficient is significant, we need to perform a
hypothesis test. The hypotheses are:
H0:β1= 0 (slope coefficient is not significant)
Ha:β1= 0 (slope coefficient is significant)
We will use the t-test statistic given by:
t=ˆ
β1
se/√∑(Xi−¯
X)2=2.6
1.4/√∑(Xi−¯
X)2
Under H0,tfollows a t-distribution with n−2 = 18 degrees of freedom. The
critical value at 5% significance level for a two-tailed test is tcritical =±2.101.
If |t|> tcritical, we reject H0; otherwise, we fail to reject H0. Let’s calculate
tand make a decision.
Step 1: Calculate the sum of squares of (Xi−¯
X):
∑(Xi−¯
X)2=∑X2
i−1
n(∑Xi)2
Step 2: Substitute the values into the formula for t:
t=2.6
1.4/√∑(Xi−¯
X)2
Step 3: Compare |t|with tcritical to make a decision.
(b) To construct a 95% confidence interval for the slope coefficient β1, we
use the formula: ˆ
β1±tα/2,n−2×se
√∑(Xi−¯
X)2
Where tα/2,n−2is the critical t-value for a 95% confidence interval with
n−2degrees of freedom. The confidence interval is calculated by substituting
the values into the formula.
Step 1: Calculate the critical t-value tα/2,n−2.
Step 2: Substitute the values into the formula to find the confidence interval
for β1.
Question 9
Question
Suppose we have collected data on a sample of 50 individuals, including their
annual income and the number of years of education they have completed. The
least squares regression line for predicting annual income based on years of
education is given by ˆ
Y= 2000 + 1000X, where Yrepresents annual income
8
(in dollars) and Xrepresents years of education. The standard error of the
estimate for this regression line is 500 dollars. Test the null hypothesis that the
true slope coefficient is equal to 0 against the alternative hypothesis that the
true slope coefficient is not equal to 0, at a significance level of 0.05.
Solution
Step 1: State the hypotheses: - Null Hypothesis (H0): The true slope coefficient
is equal to 0 (β1= 0). - Alternative Hypothesis (H1): The true slope coefficient
is not equal to 0 (β1= 0).
Step 2: Compute the test statistic: The test statistic for testing the slope
coefficient in linear regression is given by:
t=b1
SE(b1)
where b1is the estimate of the slope coefficient and SE(b1)is the standard error
of the estimate.
In this case, b1= 1000 and SE(b1) = 500. Thus, the test statistic is:
t=1000
500 = 2
Step 3: Determine the degrees of freedom: The degrees of freedom for testing
the slope coefficient in linear regression is equal to n−2, where nis the number
of observations in the sample. Here, n= 50, so the degrees of freedom is
50 −2 = 48.
Step 4: Determine the critical value: Since we are testing at a significance
level of 0.05, the critical values for a two-tailed test with 48 degrees of freedom
are approximately ±2.0106.
Step 5: Make a decision: Since the calculated test statistic of 2 does not fall
in the rejection region defined by the critical values of ±2.0106, we do not reject
the null hypothesis.
Step 6: Conclusion: We do not have enough evidence to conclude that the
true slope coefficient is significantly different from 0 at a significance level of
0.05.
Question 10
Question
Suppose we have a dataset consisting of 20 observations and 3 predictor vari-
ables. The fitted linear regression model is given by ˆ
Y= 2X1−3X2+4X3. The
residual standard error is found to be 5. Perform a hypothesis test to determine
if the coefficient of X2is significantly different from zero at the 5
9
Solution
Step 1: State the null and alternative hypotheses.
Let β2be the coefficient of X2. The hypotheses are: - Null hypothesis:
H0:β2= 0 - Alternative hypothesis: H1:β2= 0
Step 2: Calculate the t-statistic. The t-statistic is given by:
t=ˆ
β2
SE(ˆ
β2)
where ˆ
β2is the estimated coefficient of X2and SE(ˆ
β2)is the standard error
of the coefficient estimate.
Step 3: Find the p-value associated with the t-statistic. Under the null
hypothesis, the t-statistic follows a t-distribution with n−p−1 = 20−3−1 = 16
degrees of freedom (where nis the number of observations and pis the number
of predictor variables). We can find the p-value by comparing the absolute value
of the t-statistic to the critical t-value for a two-tailed test.
Step 4: Make a decision based on the p-value. At the 5
Step 5: Interpret the results. If the null hypothesis is rejected, we can
conclude that the coefficient of X2is significantly different from zero.
Question 11
Question
Let Ybe a response variable and Xbe a predictor variable in a simple linear
regression model. The data below was collected from a random sample of size
n= 10:X Y
2 5
4 7
6 10
8 14
10 13
12 16
14 19
16 20
18 21
20 25
Calculate a 95% confidence interval for the slope of the regression line.
Solution
Step 1: Calculate the sample mean of X(X), the sample mean of Y(Y), the
sample variance of X(s2
X), the sample variance of Y(s2
Y), and the sample
10
covariance between Xand Y(sXY ). Using the formulas:
X=1
n
n
∑
i=1
Xi, Y =1
n
n
∑
i=1
Yi
s2
X=1
n−1
n
∑
i=1
(Xi−X)2, s2
Y=1
n−1
n
∑
i=1
(Yi−Y)2
sXY =1
n−1
n
∑
i=1
(Xi−X)(Yi−Y)
we calculate:
X= 11, Y = 15.5
s2
X= 44.44, s2
Y= 35.33
sXY = 42.5
Step 2: Calculate the slope estimator, b1, using the formula:
b1=sXY
s2
X
Substitute the values into the formula:
b1=42.5
44.44 = 0.958
Step 3: Calculate the standard error of the slope (SEb1) using the formula:
SEb1=√s2
Y
n−2/√s2
X
Substitute the values into the formula:
SEb1=√35.33
8/√44.44 = 0.680
Step 4: Find the critical value t∗for a 95% confidence interval with n−2 = 8
degrees of freedom. From the t-distribution table, t∗= 2.306.
Step 5: Calculate the margin of error at a 95% confidence level:
ME =t∗×SEb1= 2.306 ×0.680 = 1.568
Step 6: Finally, construct the 95% confidence interval for the slope:
CI = (b1−M E, b1+ME) = (0.958 −1.568,0.958 + 1.568) = (−0.610,2.526)
Therefore, the 95% confidence interval for the slope of the regression line is
(−0.610,2.526).
11
Question 12
Question
Suppose we have a linear regression model given by Y=β0+β1X+ε, where
Yis the response variable, Xis the predictor variable, β0and β1are unknown
coefficients, and εis the error term. Given the following information from a
sample of size n= 20:
20
∑
i=1
Yi= 360,
20
∑
i=1
Xi= 30,
20
∑
i=1
Y2
i= 2440,
20
∑
i=1
X2
i= 118,
20
∑
i=1
XiYi= 450
Determine the estimated regression line of Yon Xand conduct a hypothesis
test to determine if there is a significant linear relationship between Xand Yat
the 95% confidence level. Assume all necessary assumptions for linear regression
are satisfied.
Solution
Step 1: Calculate the slope ( ˆ
β1) and intercept ( ˆ
β0) of the estimated regression
line using the formula:
ˆ
β1=n∑n
i=1 XiYi−∑n
i=1 Xi∑n
i=1 Yi
n∑n
i=1 X2
i−(∑n
i=1 Xi)2
ˆ
β0=∑n
i=1 Yi−ˆ
β1∑n
i=1 Xi
n
Substitute the given values into the formulas:
ˆ
β1=20(450) −30(360)
20(118) −302=9000 −10800
2360 −900 =−1800
1460 ≈ −1.23
ˆ
β0=360 −(−1.23)(30)
20 ≈55.6
Therefore, the estimated regression line is ˆ
Y= 55.6−1.23X.
Step 2: Perform a hypothesis test to determine if there is a significant linear
relationship between Xand Yat the 95% confidence level.
The hypotheses are: - H0:β1= 0 (There is no linear relationship between
Xand Y) - H1:β1= 0 (There is a linear relationship between Xand Y)
We will conduct a t-test using the test statistic:
t=ˆ
β1
√MSE
∑n
i=1(Xi−X)2
=−1.23
√2440−(−1.23)(450)
20−2·1
20(118)−302≈ −2.46
12
Under the null hypothesis, tfollows a t-distribution with n−2 = 18 degrees
of freedom. At the 95% confidence level, the critical value for a two-tailed test
is approximately ±2.101.
Since |t|= 2.46 >2.101, we reject the null hypothesis. Therefore, we have
enough evidence to conclude that there is a significant linear relationship be-
tween Xand Yat the 95% confidence level.
Question 14
Question
Let Yibe the response variable and Xibe the predictor variable. Consider the
following simple linear regression model:
Yi=β0+β1Xi+εi
where εi∼N(0, σ2)are independent and identically distributed. From a
sample of size n= 20, the following least squares regression line was obtained:
ˆ
Y= 5.1+2.3X. Further analysis produced the following ANOVA table:
Source Sum of Squares Degrees of Freedom Mean Square F
Regression 78.56 1 78.56 24.9
Residual 41.28 18 2.29
Total 119.84 19
Perform a hypothesis test at significance level α= 0.05 to determine whether
there is a significant linear relationship between the response variable and the
predictor variable.
Solution
Step 1: State the Hypotheses
The null hypothesis H0and the alternative hypothesis HAare: - H0:β1= 0
(There is no linear relationship between the response variable and the predictor
variable) - HA:β1= 0 (There is a linear relationship between the response
variable and the predictor variable)
Step 2: Calculate the F-statistic
The F-statistic is calculated as Mean Square Regression
Mean Square Residual =78.56
2.29 ≈34.25.
Step 3: Find the Critical Value
With degrees of freedom for the regression and residual being 1 and 18
respectively, and significance level α= 0.05, the critical value can be found using
an F-distribution table. For α= 0.05, the critical values are approximately 6.39
and 0.156.
Step 4: Make a Decision
Since the F-statistic (34.25) is greater than the critical value (6.39), we reject
the null hypothesis.
13
Step 5: State the Conclusion
There is enough evidence to conclude that there is a significant linear rela-
tionship between the response variable and the predictor variable at a signifi-
cance level of 0.05.
Question 15
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+εi, where
εiare independent identically distributed random variables with mean 0and
variance σ2.
Given the following data:
X Y
1 2
2 4
3 6
4 8
5 10
Estimate the regression coefficients β0and β1using the method of least
squares.
Solution
Step 1: We first calculate the sample means ¯
Xand ¯
Yusing the given data.
¯
X=1+2+3+4+5
5= 3 and ¯
Y=2+4+6+8+10
5= 6
Step 2: Next, we calculate the sample covariance sXY and the sample vari-
ance s2
X.
sXY =∑n
i=1(Xi−¯
X)(Yi−¯
Y)
n−1=(1 −3)(2 −6) + (2 −3)(4 −6) + (3 −3)(6 −6) + (4 −3)(8 −6) + (5 −3)(10 −6)
5−1= 4
s2
X=∑n
i=1(Xi−¯
X)2
n−1=(1 −3)2+ (2 −3)2+ (3 −3)2+ (4 −3)2+ (5 −3)2
5−1= 2.5
Step 3: Calculate the estimated regression coefficient ˆ
β1.
ˆ
β1=sXY
s2
X
=4
2.5= 1.6
Step 4: Use the estimated coefficient ˆ
β1to find the estimated intercept ˆ
β0.
ˆ
β0=¯
Y−ˆ
β1¯
X= 6 −1.6(3) = 1.8
Therefore, the estimated regression equation is ˆ
Y= 1.8+1.6X.
14
Question 16
Question
Let Yrepresent the dependent variable and Xrepresent the independent vari-
able in a simple linear regression model. A sample of 20 data points yielded the
following results:
∑Y= 400,∑X= 50,∑Y2= 16000,∑X2= 600,and ∑XY = 2500
Using these results, test at the 5
Solution
Step 1: The null and alternative hypotheses are stated as:
H0:β= 0 (slope is equal to 0)
HA:β= 0 (slope is not equal to 0)
Step 2: Calculate the sample size, the coefficient of determination R2, and
the test statistic:
n= 20
R2=∑n
i=1(ˆ
Yi−¯
Y)2
∑n
i=1(Yi−¯
Y)2=∑Y2−(∑Y)2
n
∑Y2= 1−(400
20 )2
16000
20
= 1−1002
800 = 1−10000
800 = 1−12.5 = −11.5
t=ˆ
β−0
se(ˆ
β)=
∑XY −∑X∑Y
n
∑X2−(∑X)2
n
√∑Y2−(∑Y)2
n−ˆ
β2(∑X2−(∑X)2
n)
n−2
=
2500−50·400
20
600−502
20
√16000−4002
20 −ˆ
β2(600−502
20 )
20−2
=
=2500 −2000
20
600 −2500
20
∇·√16000 −160000
20 −ˆ
β2(600 −2500
20 )
18 =2500 −100
600 −125 ∇·√16000 −8000 −ˆ
β2(600 −125)
18 =2400
475 ∇·√8000 −600 ˆ
β2
18
Step 3: Reject the null hypothesis if |t|> tn−2,α/2. Since this is a two-tailed
test at the 5
In this case, we calculate the value of |t|and compare it to the critical value.
Step 4: Calculate the p-value associated with the test statistic:
p-value = 2(1−P(tn−2≤ |t|)) = 2(1−P(t18 ≤2.097)) = 2(1−0.9756) = 2(0.0244) = 0.0488
Step 5: Make a decision: Since p-value = 0.0488 <0.05 and |t|= 2.097 >
2.101, we reject the null hypothesis.
Step 6: Conclusion: There is sufficient evidence at the 5
15
Question 17
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εi∼N(0, σ2)for i= 1,2, . . . , n. The least squares estimates for the
model coefficients are ˆ
β0= 2 and ˆ
β1= 3.5. Given that the variance of the error
terms is σ2= 4, test the hypothesis H0:β1= 3 vs Ha:β1= 3 at a significance
level of α= 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing the slope
coefficient in simple linear regression is given by:
t=ˆ
β1−β1,null
√Var(ˆ
β1)
Where β1,null is the hypothesized value of the slope coefficient under the
null hypothesis (in this case, β1,null = 3). The variance of the estimated slope
coefficient is given by:
Var(ˆ
β1) = σ2
∑n
i=1(Xi−¯
X)2
Given that σ2= 4, we need to calculate ∑n
i=1(Xi−¯
X)2. Without further
information, we are unable to calculate this directly. Therefore, we cannot
calculate the exact test statistic.
Step 2: Determine the critical value. Since we cannot calculate the exact test
statistic, we are unable to determine the critical value for the t-distribution with
n−2degrees of freedom. Hence, we are unable to proceed with the hypothesis
test without the additional information required for the calculations.
Therefore, the hypothesis test cannot be conducted without the necessary
information to calculate the test statistic and critical value accurately.
Question 18
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εiare independently and identically distributed with mean 0 and variance
σ2. A random sample of 10 observations produced the following estimated
regression equation: ˆ
Y= 12.5 + 1.6Xand a coefficient of determination R2=
0.82. Test the hypothesis H0:β1= 0 against Ha:β1= 0 at the 5% level of
significance.
16
Solution
Step 1: First, we need to find the test statistic. The test statistic is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of β1and SE(ˆ
β1)is the standard error of the estimate.
Step 2: We are given that ˆ
Y= 12.5+1.6X. From this, we can see that
ˆ
β1= 1.6.
Step 3: To find SE(ˆ
β1), we use the formula:
SE(ˆ
β1) = √MSE
SXX
where MSE is the mean squared error and SXX is the sum of squares Xi−¯
X.
Step 4: Given that R2= 0.82, we can find M SE using the formula:
MSE =(n−2) ×SSE
n−2=(10 −2) ×18.4
10 −2= 16.56
where nis the number of observations and SSE is the sum of squared errors.
Step 5: We can find SXX using the formula:
SXX =∑(Xi−¯
X)2= 122.5
where ¯
Xis the mean of the Xivalues.
Step 6: Substituting the values into the formula, we get:
SE(ˆ
β1) = √16.56
122.5≈0.3888
Step 7: Now, we can find the test statistic:
t=1.6
0.3888 ≈4.1171
Step 8: Since this is a two-tailed test, we will compare this value with the
critical value at the 5
Step 9: The critical values are approximately ±2.306.
Step 10: Since |4.1171|>2.306, we reject the null hypothesis H0:β1= 0.
Therefore, we have sufficient evidence to conclude that β1is not equal to 0 at
the 5% significance level.
Question 19
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ε,
where ε∼N(0, σ2).
For a specific dataset, we obtain the following results:
17
Coefficient Estimate Standard Error
β04.2 0.6
β12.8 0.4
Perform a hypothesis test to determine if there is a significant linear rela-
tionship between Xand Yat a 5
Solution
Step 1: Formulate the hypotheses
The null hypothesis is that there is no linear relationship between Xand Y:
H0:β1= 0
The alternative hypothesis is that there is a significant linear relationship be-
tween Xand Y:
Ha:β1= 0
Step 2: Calculate the test statistic
The test statistic for testing the slope coefficient β1is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of coefficient β1and SE(ˆ
β1)is the standard error of
the estimate.
In this case,
t=2.8
0.4= 7
Step 3: Determine the critical value
Since the alternative hypothesis is two-sided, we will look for the critical values
in the t-distribution table for a two-tailed test with α= 0.05 and degrees of
freedom n−2(for simple linear regression).
For a significance level of 5
Step 4: Make a decision
Since the test statistic t = 7 is greater than 2.306, we reject the null hypothesis.
Therefore, we have enough evidence to conclude that there is a significant
linear relationship between Xand Y.
Question 20
Question
Suppose a linear regression model is fitted to a set of data and the estimated
regression equation is ˆy= 3.2 + 2.5x. The standard error of the estimate is
4.8. Test the null hypothesis H0:β1= 0 against the alternative hypothesis
Ha:β1= 0 at the 5
18
Solution
Step 1: The test statistic tis calculated using the formula:
t=b1
SE(b1)
where b1is the estimated coefficient for variable x, and SE(b1)is the standard
error of this coefficient.
Step 2: Given that the estimated coefficient b1= 2.5and the standard error
SE(b1) = 4.8, we can substitute these values into the formula to find the test
statistic:
t=2.5
4.8
Step 3: Calculate the value of t:
t≈0.52
Step 4: Degrees of freedom for the test statistic tis equal to the sample size
minus 2, where 2 is the number of parameters estimated in the regression model
(intercept and slope).
df =n−2
Step 5: Since the significance level is 5
Step 6: Compare the calculated tvalue with the critical values. Since 0.52
falls within the range (-2.571, 2.571), we fail to reject the null hypothesis H0:
β1= 0.
Step 7: Therefore, at 5
Question 21
Question
Let Yidenote the response variable and Xidenote the predictor variable for a
linear regression model. Consider the following multiple linear regression model:
Yi=β0+β1X1i+β2X2i+εi
where εi∼N(0, σ2)are independent and identically distributed.
Given the data set (X1i, X2i, Yi)for i= 1,2,3, . . . , n, we are interested in
testing the null hypothesis H0:β1=β2against the alternative hypothesis
Ha:β1=β2using a significance level of α= 0.05. The fitted model results in
ˆ
β1= 3.9,ˆ
β2= 4.1, and the standard error of ˆ
β1is 0.6 while the standard error
of ˆ
β2is 0.5.
Perform a hypothesis test for β1=β2using the given information.
19
Solution
Step 1: Formulate the test statistic. Given that we have to test whether β1=β2,
the test statistic is computed as
t=ˆ
β1−ˆ
β2
√SE(ˆ
β1)2+SE(ˆ
β2)2
Step 2: Compute the test statistic. Plugging in the given values, we obtain
t=3.9−4.1
√0.62+ 0.52
Step 3: Determine the critical value. Since we are performing a two-tailed
test at a significance level of α= 0.05, the critical values are ±zα/2where
zα/2=z0.025 ≈1.96.
Step 4: Make a decision. If |t|>1.96, we reject the null hypothesis H0:
β1=β2. Otherwise, we fail to reject H0.
Step 5: Calculate the test statistic value. Now, substitute the values and
compute tto determine the decision:
t=3.9−4.1
√0.62+ 0.52=−0.2
√0.36 + 0.25 ≈−0.2
√0.61 ≈ −0.255
Since | − 0.255|<1.96, we fail to reject H0.
Step 6: Formulate the conclusion. Since the test statistic value falls within
the range of the critical values, we do not have enough evidence to conclude
that β1is different from β2.
Question 22
Question
Let Y1, Y2, . . . , Ynbe independent and identically distributed random variables
following a normal distribution with mean µand variance σ2. Consider the
simple linear regression model Yi=β0+β1Xi+εi, where β0,β1, and σ2are
unknown parameters to be estimated.
Suppose that the least squares estimates of β0and β1are denoted as ˆ
β0and
ˆ
β1, respectively. Show that ˆ
β0and ˆ
β1are unbiased estimators for β0and β1,
respectively.
Solution
Step 1: Recall that the least squares estimates of β0and β1are given by:
ˆ
β1=
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
20
ˆ
β0=¯
Y−ˆ
β1¯
X
Step 2: Let’s start by showing that ˆ
β1is an unbiased estimator for β1. The
expectation of ˆ
β1can be computed as follows:
E(ˆ
β1) = E
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)E(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−E(¯
Y))
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−µ)
n
∑
i=1
(Xi−¯
X)2
Step 3: Since the Yi’s are independent and identically distributed with mean
µ, we have E(Yi) = µfor all i. Thus, E(ˆ
β1) =
n
∑
i=1
(Xi−¯
X)·0
n
∑
i=1
(Xi−¯
X)2= 0.
Step 4: Therefore, ˆ
β1is an unbiased estimator for β1.
Step 5: Next, we will show that ˆ
β0is an unbiased estimator for β0. By
linearity of expectation, we have:
E(ˆ
β0) = E(¯
Y)−E(ˆ
β1¯
X)
Step 6: Since ¯
Yand ¯
Xare constants, we can simplify this expression to:
E(ˆ
β0) = ¯
Y−¯
XE(ˆ
β1)
Step 7: Since ˆ
β1is an unbiased estimator for β1, we have E(ˆ
β1) = 0. Thus,
E(ˆ
β0) = ¯
Y−0·¯
X=¯
Y.
Step 8: Therefore, ˆ
β0is an unbiased estimator for β0.
Question 23
Question
Suppose you are given the following multiple regression model:
y= 3x1−2x2+ 5x3+ 4
21
where yis the dependent variable, and x1,x2, and x3are the independent
variables. You are also given the following outputs from the regression analysis:
• Residual standard error = 2.5
•R2= 0.85
• F-statistic = 15.6
Is this model statistically significant at the α= 0.05 level? Justify your answer.
Solution
Step 1: Hypotheses
The null hypothesis H0is that all the coefficients in the regression model are
equal to zero:
H0:β1=β2=β3= 0
The alternative hypothesis H1is that at least one coefficient is not equal to
zero:
H1: At least one of β1,β2,β3is not equal to zero.
Step 2: F-statistic test
First, we calculate the degrees of freedom for the numerator and denominator
of the F-statistic:
• Numerator degrees of freedom = number of coefficients being tested = 3
• Denominator degrees of freedom = number of observations - number of
coefficients = n−k−1, where nis the number of observations and kis
the number of coefficients including the intercept term. Since we do not
have the sample size, we cannot perform the hypothesis test.
Step 3: Decision
We cannot perform the F-test without knowing the sample size n. Therefore, we
are unable to determine the statistical significance of the model at the α= 0.05
level without additional information.
Question 24
Question
Let’s consider a multiple linear regression model with three predictors X1,X2,
and X3. The estimated regression equation is given by:
ˆ
Y= 15 + 3X1−2X2+ 4X3
Assume the estimated variances for the coefficient estimates are V ar(ˆ
β0) = 10,
V ar(ˆ
β1) = 4,V ar(ˆ
β2) = 6, and V ar(ˆ
β3) = 8. Suppose we want to test the
following null hypothesis:
H0:β1= 0, H1:β1= 0
22
Perform a hypothesis test to determine whether there is a linear relationship
between X1and Yusing a significance level of 0.05.
Solution
Step 1: The test statistic for testing the hypothesis βj= 0 versus Ha:βj= 0
is given by:
t=ˆ
βj
√V ar(ˆ
βj)
Step 2: In this case, we are interested in testing the hypothesis H0:β1= 0
versus H1:β1= 0. The test statistic for β1is:
t=3
√4=3
2
Step 3: The critical value for a two-tailed test at a 5
Step 4: Using a statistical table, we find that t0.025,16 ≈2.12.
Step 5: Compare the absolute value of the test statistic to the critical value.
Since 3
2= 1.5<2.12, we do not reject the null hypothesis.
Step 6: Therefore, we do not have enough evidence to conclude that there
is a linear relationship between X1and Yat a significance level of 0.05.
Question 25
Question
Let Ybe the dependent variable and Xbe the independent variable in a
linear regression model. Suppose that the estimated regression equation is
ˆ
Y= 3 + 1.5X. The standard error of the estimate is found to be 2.5. Given
this information, test the null hypothesis H0:β1= 0 against the alternative
hypothesis Ha:β1= 0 at the 5
Solution
Step 1: Determine the test statistic using the formula:
t=ˆ
β1−0
SE(ˆ
β1)
where ˆ
β1= 1.5and SE(ˆ
β1) = 2.5. Plugging in these values, we get:
t=1.5−0
2.5= 0.6
Step 2: Find the degrees of freedom. Since we are testing the slope coeffi-
cient, the degrees of freedom is n−2, where nis the sample size. However, the
sample size is not given in the question.
23
Step 4: Calculate the test statistic value. We have SSR =β2
1×SSXX and
SSE =SSE/df where SSXX is the sum of squares of Xand df is the degrees of
freedom for the error.
Plugging in the given values, we find:
SSR = 1.52×SSXX
SSE = 4.0/18
Step 5: Determine the critical value. Look up the critical value from the
F-distribution table with degrees of freedom 1 and 18 at the 5% significance
level. Let’s say the critical value is Fcrit.
Step 6: Make a decision. If F > Fcrit, reject the null hypothesis. Otherwise,
fail to reject the null hypothesis.
Step 7: Conclusion. State your conclusion based on the decision made in
step 6.
Question 2
Question
Suppose we have a dataset with the following linear regression model:
Y= 3X1−2X2+ 5X3+ϵ
Where X1, X2, and X3are independent variables, and ϵis the error term. Af-
ter performing linear regression analysis, we obtained the following estimated
regression coefficients:
ˆ
β1= 2.6,ˆ
β2=−1.8,ˆ
β3= 4.9
Also, the standard errors of these estimates were found to be:
SE(ˆ
β1) = 0.3, SE(ˆ
β2) = 0.4, SE(ˆ
β3) = 0.2
Using a significance level of 0.05, test the null hypothesis that the coefficient for
X2is equal to zero.
Solution
Step 1: Formulate the hypotheses.
The null hypothesis (H0) and alternative hypothesis (H1) can be formulated as:
H0:β2= 0
H1:β2= 0
2
Step 2: Calculate the test statistic.
The test statistic for testing the significance of β2is given by:
t=ˆ
β2
SE(ˆ
β2)
Substitute the given values to find the test statistic:
t=−1.8
0.4=−4.5
Step 3: Determine the degrees of freedom.
The degrees of freedom for the t-distribution in this case is n−3, where nis the
number of observations. Since the degrees of freedom is not explicitly given in
the question, we assume that there are enough observations to ensure that the
t-distribution is a good approximation.
Step 4: Determine the critical value.
Using a significance level of 0.05 and the two-tailed test, the critical value is
found using a t-table with appropriate degrees of freedom.
Step 5: Make a decision.
Compare the absolute value of the test statistic with the critical value from the
t-table. If the absolute value of the test statistic is greater than the critical value,
we reject the null hypothesis; otherwise, we fail to reject the null hypothesis.
Step 6: Conclusion.
Based on the decision in Step 5, make a conclusion regarding the null hypothesis
at the 0.05 significance level.
Question 4
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+ϵi, where
ϵi∼N(0, σ2)represents the error term. We are interested in testing the null
hypothesis H0:β1= 0 against the alternative hypothesis H1:β1= 0. Given
the following data and output summary from a regression analysis:
Variable ˆ
βSE(ˆ
β)
β03.56 0.42
β10.78 0.06
Perform the hypothesis test using a significance level of 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing whether the
slope is significantly different from zero is given by:
t=ˆ
β1−0
SE(ˆ
β1)
3
Given ˆ
β1= 0.78 and SE(ˆ
β1) = 0.06, we have:
t=0.78 −0
0.06 = 13
Step 2: Determine the critical value. Since we are testing for a two-tailed
hypothesis test at a significance level of 0.05, the critical value can be obtained
using a t-distribution with n−2degrees of freedom.
For a two-tailed test with α= 0.05 and n−2 = 10 −2=8degrees of
freedom, the critical values are ±2.306.
Step 3: Make a decision. Since |t|= 13 >2.306, we reject the null hypothesis
H0:β1= 0 in favor of the alternative hypothesis H1:β1= 0 at the 0.05
significance level. This indicates that the slope coefficient β1is significantly
different from zero.
Question 5
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where Yrepresents the dependent variable, Xthe independent variable, and
ϵthe error term. After conducting a hypothesis test, we obtain the following
results: the test statistic is t= 2.76 and the critical value at the 5
Given this information, can we reject the null hypothesis at a significance
level of 5
Solution
Step 1: State the null and alternative hypotheses.
The null hypothesis states that the slope coefficient is equal to 0 (H0:β1= 0),
while the alternative hypothesis states that the slope coefficient is not equal to
0 (Ha:β1= 0).
Step 2: Determine the decision rule.
Since the test statistic t= 2.76, and the critical value at the 5
Step 3: Make a decision.
Since |2.76|>2.262, we reject the null hypothesis.
Step 4: State the conclusion.
Therefore, at the 5
Question 6
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ϵ,
where the error term ϵfollows a normal distribution with mean 0 and variance
4
σ2. In a study examining the relationship between hours studied (indepen-
dent variable) and exam score (dependent variable), a researcher obtained the
following results for the regression analysis:
ˆ
β0= 60.5,ˆ
β1= 3.2,
s2= 9.6,
n= 30.
Determine a 95% confidence interval for the true slope β1.
Solution
Step 1: The formula to compute the confidence interval for the slope estimate
β1is given by:
ˆ
β1±tα/2,n−2·s
√∑n
i=1(Xi−¯
X)2
,
where tα/2,n−2is the critical value from the t-distribution with n−2degrees of
freedom.
Step 2: First, we compute the critical value tα/2,n−2at the 95% confidence
level. Since α= 0.05, the degrees of freedom are n−2 = 30 −2 = 28. Using a
t-table or a statistical software, we find t0.025,28 ≈2.048.
Step 3: Next, we calculate the standard error of the estimate:
SE(ˆ
β1) = s
√∑n
i=1(Xi−¯
X)2
.
Step 4: Given s2= 9.6, we have s=√9.6 = 3.1. Since we do not have
the individual data values, we cannot directly compute the sum of squares of
(Xi−¯
X). However, we can use the fact that ∑n
i=1(Xi−¯
X)2= (n−1) ·s2
X,
where s2
Xis the sample variance of the independent variable.
Step 5: We can estimate s2
Xusing s2, which gives s2
X=s2
n−1=9.6
30−1≈0.32.
Therefore, the standard error of the estimate is SE(ˆ
β1) = 3.1
√0.32 ≈1.75.
Step 6: Finally, we can calculate the confidence interval by plugging into the
formula:
3.2±2.048 ×1.75.
Step 7: Thus, the 95% confidence interval for the true slope parameter β1is
approximately [0.595,5.805].
Question 7
Question
Let Ybe a random variable representing the response variable and Xbe a
random variable representing the predictor variable in a linear regression model.
5
Suppose we have the following simple linear regression model:
Y=β0+β1X+ε,
where εis the error term with mean 0 and variance σ2. Given a sample of size
n= 20, the following data was obtained:
X Y
1 2.3
2 4.5
3 6.7
4 8.9
5 11.1
6 13.3
7 15.5
8 17.7
9 19.9
10 22.1
11 24.3
12 26.5
13 28.7
14 30.9
15 33.1
16 35.3
17 37.5
18 39.7
19 41.9
20 44.1
Determine the least squares estimators ˆ
β0and ˆ
β1, the standard errors of
these estimators, and the 95
Solution
Step 1: Compute the least squares estimators ˆ
β0and ˆ
β1using the formulas:
ˆ
β1=∑n
i=1(Xi−¯
X)(Yi−¯
Y)
∑n
i=1(Xi−¯
X)2,
ˆ
β0=¯
Y−ˆ
β1¯
X,
where ¯
Xand ¯
Yare the sample means of Xand Yrespectively.
Step 2: Calculate the sample means:
¯
X=1
n
n
∑
i=1
Xi= 10.5,
6
¯
Y=1
n
n
∑
i=1
Yi= 22.2.
Step 3: Compute the least squares estimators:
ˆ
β1=∑20
i=1(Xi−10.5)(Yi−22.2)
∑20
i=1(Xi−10.5)2=203.4
215 ≈0.94651,
ˆ
β0= 22.2−0.94651 ×10.5≈11.49355.
Step 4: Compute the residuals ˆεi=Yi−ˆ
β0−ˆ
β1Xifor each observation.
Step 5: Calculate the standard errors of ˆ
β0and ˆ
β1using the formulas:
SE(ˆ
β0) = ˆσ√1
n+¯
X2
∑n
i=1(Xi−¯
X)2,
SE(ˆ
β1) = ˆσ
√∑n
i=1(Xi−¯
X)2
,
where ˆσis the estimate of the standard deviation of the error term.
Step 6: Finally, compute the 95
ˆ
β0−tα/2,n−2×SE(ˆ
β0)≤β0≤ˆ
β0+tα/2,n−2×SE(ˆ
β0),
ˆ
β1−tα/2,n−2×SE
Question 8
Question
Let Yidenote the response variable and Xibe the predictor variable for i=
1,2, . . . , n observations. Consider the simple linear regression model given by
Yi=β0+β1Xi+εiwhere εiare independent and identically distributed with
mean zero and variance σ2. Suppose we have fitted the model and obtained the
following output:
ˆ
Y= 4.2+2.6X, se= 1.4, n = 20
(a) Test at the 5% significance level whether the slope coefficient is signifi-
cant.
(b) Construct a 95% confidence interval for the slope coefficient.
7
Solution
(a) To test whether the slope coefficient is significant, we need to perform a
hypothesis test. The hypotheses are:
H0:β1= 0 (slope coefficient is not significant)
Ha:β1= 0 (slope coefficient is significant)
We will use the t-test statistic given by:
t=ˆ
β1
se/√∑(Xi−¯
X)2=2.6
1.4/√∑(Xi−¯
X)2
Under H0,tfollows a t-distribution with n−2 = 18 degrees of freedom. The
critical value at 5% significance level for a two-tailed test is tcritical =±2.101.
If |t|> tcritical, we reject H0; otherwise, we fail to reject H0. Let’s calculate
tand make a decision.
Step 1: Calculate the sum of squares of (Xi−¯
X):
∑(Xi−¯
X)2=∑X2
i−1
n(∑Xi)2
Step 2: Substitute the values into the formula for t:
t=2.6
1.4/√∑(Xi−¯
X)2
Step 3: Compare |t|with tcritical to make a decision.
(b) To construct a 95% confidence interval for the slope coefficient β1, we
use the formula: ˆ
β1±tα/2,n−2×se
√∑(Xi−¯
X)2
Where tα/2,n−2is the critical t-value for a 95% confidence interval with
n−2degrees of freedom. The confidence interval is calculated by substituting
the values into the formula.
Step 1: Calculate the critical t-value tα/2,n−2.
Step 2: Substitute the values into the formula to find the confidence interval
for β1.
Question 9
Question
Suppose we have collected data on a sample of 50 individuals, including their
annual income and the number of years of education they have completed. The
least squares regression line for predicting annual income based on years of
education is given by ˆ
Y= 2000 + 1000X, where Yrepresents annual income
8
(in dollars) and Xrepresents years of education. The standard error of the
estimate for this regression line is 500 dollars. Test the null hypothesis that the
true slope coefficient is equal to 0 against the alternative hypothesis that the
true slope coefficient is not equal to 0, at a significance level of 0.05.
Solution
Step 1: State the hypotheses: - Null Hypothesis (H0): The true slope coefficient
is equal to 0 (β1= 0). - Alternative Hypothesis (H1): The true slope coefficient
is not equal to 0 (β1= 0).
Step 2: Compute the test statistic: The test statistic for testing the slope
coefficient in linear regression is given by:
t=b1
SE(b1)
where b1is the estimate of the slope coefficient and SE(b1)is the standard error
of the estimate.
In this case, b1= 1000 and SE(b1) = 500. Thus, the test statistic is:
t=1000
500 = 2
Step 3: Determine the degrees of freedom: The degrees of freedom for testing
the slope coefficient in linear regression is equal to n−2, where nis the number
of observations in the sample. Here, n= 50, so the degrees of freedom is
50 −2 = 48.
Step 4: Determine the critical value: Since we are testing at a significance
level of 0.05, the critical values for a two-tailed test with 48 degrees of freedom
are approximately ±2.0106.
Step 5: Make a decision: Since the calculated test statistic of 2 does not fall
in the rejection region defined by the critical values of ±2.0106, we do not reject
the null hypothesis.
Step 6: Conclusion: We do not have enough evidence to conclude that the
true slope coefficient is significantly different from 0 at a significance level of
0.05.
Question 10
Question
Suppose we have a dataset consisting of 20 observations and 3 predictor vari-
ables. The fitted linear regression model is given by ˆ
Y= 2X1−3X2+4X3. The
residual standard error is found to be 5. Perform a hypothesis test to determine
if the coefficient of X2is significantly different from zero at the 5
9
Solution
Step 1: State the null and alternative hypotheses.
Let β2be the coefficient of X2. The hypotheses are: - Null hypothesis:
H0:β2= 0 - Alternative hypothesis: H1:β2= 0
Step 2: Calculate the t-statistic. The t-statistic is given by:
t=ˆ
β2
SE(ˆ
β2)
where ˆ
β2is the estimated coefficient of X2and SE(ˆ
β2)is the standard error
of the coefficient estimate.
Step 3: Find the p-value associated with the t-statistic. Under the null
hypothesis, the t-statistic follows a t-distribution with n−p−1 = 20−3−1 = 16
degrees of freedom (where nis the number of observations and pis the number
of predictor variables). We can find the p-value by comparing the absolute value
of the t-statistic to the critical t-value for a two-tailed test.
Step 4: Make a decision based on the p-value. At the 5
Step 5: Interpret the results. If the null hypothesis is rejected, we can
conclude that the coefficient of X2is significantly different from zero.
Question 11
Question
Let Ybe a response variable and Xbe a predictor variable in a simple linear
regression model. The data below was collected from a random sample of size
n= 10:X Y
2 5
4 7
6 10
8 14
10 13
12 16
14 19
16 20
18 21
20 25
Calculate a 95% confidence interval for the slope of the regression line.
Solution
Step 1: Calculate the sample mean of X(X), the sample mean of Y(Y), the
sample variance of X(s2
X), the sample variance of Y(s2
Y), and the sample
10
covariance between Xand Y(sXY ). Using the formulas:
X=1
n
n
∑
i=1
Xi, Y =1
n
n
∑
i=1
Yi
s2
X=1
n−1
n
∑
i=1
(Xi−X)2, s2
Y=1
n−1
n
∑
i=1
(Yi−Y)2
sXY =1
n−1
n
∑
i=1
(Xi−X)(Yi−Y)
we calculate:
X= 11, Y = 15.5
s2
X= 44.44, s2
Y= 35.33
sXY = 42.5
Step 2: Calculate the slope estimator, b1, using the formula:
b1=sXY
s2
X
Substitute the values into the formula:
b1=42.5
44.44 = 0.958
Step 3: Calculate the standard error of the slope (SEb1) using the formula:
SEb1=√s2
Y
n−2/√s2
X
Substitute the values into the formula:
SEb1=√35.33
8/√44.44 = 0.680
Step 4: Find the critical value t∗for a 95% confidence interval with n−2 = 8
degrees of freedom. From the t-distribution table, t∗= 2.306.
Step 5: Calculate the margin of error at a 95% confidence level:
ME =t∗×SEb1= 2.306 ×0.680 = 1.568
Step 6: Finally, construct the 95% confidence interval for the slope:
CI = (b1−M E, b1+ME) = (0.958 −1.568,0.958 + 1.568) = (−0.610,2.526)
Therefore, the 95% confidence interval for the slope of the regression line is
(−0.610,2.526).
11
Question 12
Question
Suppose we have a linear regression model given by Y=β0+β1X+ε, where
Yis the response variable, Xis the predictor variable, β0and β1are unknown
coefficients, and εis the error term. Given the following information from a
sample of size n= 20:
20
∑
i=1
Yi= 360,
20
∑
i=1
Xi= 30,
20
∑
i=1
Y2
i= 2440,
20
∑
i=1
X2
i= 118,
20
∑
i=1
XiYi= 450
Determine the estimated regression line of Yon Xand conduct a hypothesis
test to determine if there is a significant linear relationship between Xand Yat
the 95% confidence level. Assume all necessary assumptions for linear regression
are satisfied.
Solution
Step 1: Calculate the slope ( ˆ
β1) and intercept ( ˆ
β0) of the estimated regression
line using the formula:
ˆ
β1=n∑n
i=1 XiYi−∑n
i=1 Xi∑n
i=1 Yi
n∑n
i=1 X2
i−(∑n
i=1 Xi)2
ˆ
β0=∑n
i=1 Yi−ˆ
β1∑n
i=1 Xi
n
Substitute the given values into the formulas:
ˆ
β1=20(450) −30(360)
20(118) −302=9000 −10800
2360 −900 =−1800
1460 ≈ −1.23
ˆ
β0=360 −(−1.23)(30)
20 ≈55.6
Therefore, the estimated regression line is ˆ
Y= 55.6−1.23X.
Step 2: Perform a hypothesis test to determine if there is a significant linear
relationship between Xand Yat the 95% confidence level.
The hypotheses are: - H0:β1= 0 (There is no linear relationship between
Xand Y) - H1:β1= 0 (There is a linear relationship between Xand Y)
We will conduct a t-test using the test statistic:
t=ˆ
β1
√MSE
∑n
i=1(Xi−X)2
=−1.23
√2440−(−1.23)(450)
20−2·1
20(118)−302≈ −2.46
12
Under the null hypothesis, tfollows a t-distribution with n−2 = 18 degrees
of freedom. At the 95% confidence level, the critical value for a two-tailed test
is approximately ±2.101.
Since |t|= 2.46 >2.101, we reject the null hypothesis. Therefore, we have
enough evidence to conclude that there is a significant linear relationship be-
tween Xand Yat the 95% confidence level.
Question 14
Question
Let Yibe the response variable and Xibe the predictor variable. Consider the
following simple linear regression model:
Yi=β0+β1Xi+εi
where εi∼N(0, σ2)are independent and identically distributed. From a
sample of size n= 20, the following least squares regression line was obtained:
ˆ
Y= 5.1+2.3X. Further analysis produced the following ANOVA table:
Source Sum of Squares Degrees of Freedom Mean Square F
Regression 78.56 1 78.56 24.9
Residual 41.28 18 2.29
Total 119.84 19
Perform a hypothesis test at significance level α= 0.05 to determine whether
there is a significant linear relationship between the response variable and the
predictor variable.
Solution
Step 1: State the Hypotheses
The null hypothesis H0and the alternative hypothesis HAare: - H0:β1= 0
(There is no linear relationship between the response variable and the predictor
variable) - HA:β1= 0 (There is a linear relationship between the response
variable and the predictor variable)
Step 2: Calculate the F-statistic
The F-statistic is calculated as Mean Square Regression
Mean Square Residual =78.56
2.29 ≈34.25.
Step 3: Find the Critical Value
With degrees of freedom for the regression and residual being 1 and 18
respectively, and significance level α= 0.05, the critical value can be found using
an F-distribution table. For α= 0.05, the critical values are approximately 6.39
and 0.156.
Step 4: Make a Decision
Since the F-statistic (34.25) is greater than the critical value (6.39), we reject
the null hypothesis.
13
Step 5: State the Conclusion
There is enough evidence to conclude that there is a significant linear rela-
tionship between the response variable and the predictor variable at a signifi-
cance level of 0.05.
Question 15
Question
Suppose we have a linear regression model given by Yi=β0+β1Xi+εi, where
εiare independent identically distributed random variables with mean 0and
variance σ2.
Given the following data:
X Y
1 2
2 4
3 6
4 8
5 10
Estimate the regression coefficients β0and β1using the method of least
squares.
Solution
Step 1: We first calculate the sample means ¯
Xand ¯
Yusing the given data.
¯
X=1+2+3+4+5
5= 3 and ¯
Y=2+4+6+8+10
5= 6
Step 2: Next, we calculate the sample covariance sXY and the sample vari-
ance s2
X.
sXY =∑n
i=1(Xi−¯
X)(Yi−¯
Y)
n−1=(1 −3)(2 −6) + (2 −3)(4 −6) + (3 −3)(6 −6) + (4 −3)(8 −6) + (5 −3)(10 −6)
5−1= 4
s2
X=∑n
i=1(Xi−¯
X)2
n−1=(1 −3)2+ (2 −3)2+ (3 −3)2+ (4 −3)2+ (5 −3)2
5−1= 2.5
Step 3: Calculate the estimated regression coefficient ˆ
β1.
ˆ
β1=sXY
s2
X
=4
2.5= 1.6
Step 4: Use the estimated coefficient ˆ
β1to find the estimated intercept ˆ
β0.
ˆ
β0=¯
Y−ˆ
β1¯
X= 6 −1.6(3) = 1.8
Therefore, the estimated regression equation is ˆ
Y= 1.8+1.6X.
14
Question 16
Question
Let Yrepresent the dependent variable and Xrepresent the independent vari-
able in a simple linear regression model. A sample of 20 data points yielded the
following results:
∑Y= 400,∑X= 50,∑Y2= 16000,∑X2= 600,and ∑XY = 2500
Using these results, test at the 5
Solution
Step 1: The null and alternative hypotheses are stated as:
H0:β= 0 (slope is equal to 0)
HA:β= 0 (slope is not equal to 0)
Step 2: Calculate the sample size, the coefficient of determination R2, and
the test statistic:
n= 20
R2=∑n
i=1(ˆ
Yi−¯
Y)2
∑n
i=1(Yi−¯
Y)2=∑Y2−(∑Y)2
n
∑Y2= 1−(400
20 )2
16000
20
= 1−1002
800 = 1−10000
800 = 1−12.5 = −11.5
t=ˆ
β−0
se(ˆ
β)=
∑XY −∑X∑Y
n
∑X2−(∑X)2
n
√∑Y2−(∑Y)2
n−ˆ
β2(∑X2−(∑X)2
n)
n−2
=
2500−50·400
20
600−502
20
√16000−4002
20 −ˆ
β2(600−502
20 )
20−2
=
=2500 −2000
20
600 −2500
20
∇·√16000 −160000
20 −ˆ
β2(600 −2500
20 )
18 =2500 −100
600 −125 ∇·√16000 −8000 −ˆ
β2(600 −125)
18 =2400
475 ∇·√8000 −600 ˆ
β2
18
Step 3: Reject the null hypothesis if |t|> tn−2,α/2. Since this is a two-tailed
test at the 5
In this case, we calculate the value of |t|and compare it to the critical value.
Step 4: Calculate the p-value associated with the test statistic:
p-value = 2(1−P(tn−2≤ |t|)) = 2(1−P(t18 ≤2.097)) = 2(1−0.9756) = 2(0.0244) = 0.0488
Step 5: Make a decision: Since p-value = 0.0488 <0.05 and |t|= 2.097 >
2.101, we reject the null hypothesis.
Step 6: Conclusion: There is sufficient evidence at the 5
15
Question 17
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εi∼N(0, σ2)for i= 1,2, . . . , n. The least squares estimates for the
model coefficients are ˆ
β0= 2 and ˆ
β1= 3.5. Given that the variance of the error
terms is σ2= 4, test the hypothesis H0:β1= 3 vs Ha:β1= 3 at a significance
level of α= 0.05.
Solution
Step 1: Calculate the test statistic. The test statistic for testing the slope
coefficient in simple linear regression is given by:
t=ˆ
β1−β1,null
√Var(ˆ
β1)
Where β1,null is the hypothesized value of the slope coefficient under the
null hypothesis (in this case, β1,null = 3). The variance of the estimated slope
coefficient is given by:
Var(ˆ
β1) = σ2
∑n
i=1(Xi−¯
X)2
Given that σ2= 4, we need to calculate ∑n
i=1(Xi−¯
X)2. Without further
information, we are unable to calculate this directly. Therefore, we cannot
calculate the exact test statistic.
Step 2: Determine the critical value. Since we cannot calculate the exact test
statistic, we are unable to determine the critical value for the t-distribution with
n−2degrees of freedom. Hence, we are unable to proceed with the hypothesis
test without the additional information required for the calculations.
Therefore, the hypothesis test cannot be conducted without the necessary
information to calculate the test statistic and critical value accurately.
Question 18
Question
Suppose we have a simple linear regression model given by Yi=β0+β1Xi+εi,
where εiare independently and identically distributed with mean 0 and variance
σ2. A random sample of 10 observations produced the following estimated
regression equation: ˆ
Y= 12.5 + 1.6Xand a coefficient of determination R2=
0.82. Test the hypothesis H0:β1= 0 against Ha:β1= 0 at the 5% level of
significance.
16
Solution
Step 1: First, we need to find the test statistic. The test statistic is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of β1and SE(ˆ
β1)is the standard error of the estimate.
Step 2: We are given that ˆ
Y= 12.5+1.6X. From this, we can see that
ˆ
β1= 1.6.
Step 3: To find SE(ˆ
β1), we use the formula:
SE(ˆ
β1) = √MSE
SXX
where MSE is the mean squared error and SXX is the sum of squares Xi−¯
X.
Step 4: Given that R2= 0.82, we can find M SE using the formula:
MSE =(n−2) ×SSE
n−2=(10 −2) ×18.4
10 −2= 16.56
where nis the number of observations and SSE is the sum of squared errors.
Step 5: We can find SXX using the formula:
SXX =∑(Xi−¯
X)2= 122.5
where ¯
Xis the mean of the Xivalues.
Step 6: Substituting the values into the formula, we get:
SE(ˆ
β1) = √16.56
122.5≈0.3888
Step 7: Now, we can find the test statistic:
t=1.6
0.3888 ≈4.1171
Step 8: Since this is a two-tailed test, we will compare this value with the
critical value at the 5
Step 9: The critical values are approximately ±2.306.
Step 10: Since |4.1171|>2.306, we reject the null hypothesis H0:β1= 0.
Therefore, we have sufficient evidence to conclude that β1is not equal to 0 at
the 5% significance level.
Question 19
Question
Suppose we have a simple linear regression model given by Y=β0+β1X+ε,
where ε∼N(0, σ2).
For a specific dataset, we obtain the following results:
17
Coefficient Estimate Standard Error
β04.2 0.6
β12.8 0.4
Perform a hypothesis test to determine if there is a significant linear rela-
tionship between Xand Yat a 5
Solution
Step 1: Formulate the hypotheses
The null hypothesis is that there is no linear relationship between Xand Y:
H0:β1= 0
The alternative hypothesis is that there is a significant linear relationship be-
tween Xand Y:
Ha:β1= 0
Step 2: Calculate the test statistic
The test statistic for testing the slope coefficient β1is given by:
t=ˆ
β1
SE(ˆ
β1)
where ˆ
β1is the estimate of coefficient β1and SE(ˆ
β1)is the standard error of
the estimate.
In this case,
t=2.8
0.4= 7
Step 3: Determine the critical value
Since the alternative hypothesis is two-sided, we will look for the critical values
in the t-distribution table for a two-tailed test with α= 0.05 and degrees of
freedom n−2(for simple linear regression).
For a significance level of 5
Step 4: Make a decision
Since the test statistic t = 7 is greater than 2.306, we reject the null hypothesis.
Therefore, we have enough evidence to conclude that there is a significant
linear relationship between Xand Y.
Question 20
Question
Suppose a linear regression model is fitted to a set of data and the estimated
regression equation is ˆy= 3.2 + 2.5x. The standard error of the estimate is
4.8. Test the null hypothesis H0:β1= 0 against the alternative hypothesis
Ha:β1= 0 at the 5
18
Solution
Step 1: The test statistic tis calculated using the formula:
t=b1
SE(b1)
where b1is the estimated coefficient for variable x, and SE(b1)is the standard
error of this coefficient.
Step 2: Given that the estimated coefficient b1= 2.5and the standard error
SE(b1) = 4.8, we can substitute these values into the formula to find the test
statistic:
t=2.5
4.8
Step 3: Calculate the value of t:
t≈0.52
Step 4: Degrees of freedom for the test statistic tis equal to the sample size
minus 2, where 2 is the number of parameters estimated in the regression model
(intercept and slope).
df =n−2
Step 5: Since the significance level is 5
Step 6: Compare the calculated tvalue with the critical values. Since 0.52
falls within the range (-2.571, 2.571), we fail to reject the null hypothesis H0:
β1= 0.
Step 7: Therefore, at 5
Question 21
Question
Let Yidenote the response variable and Xidenote the predictor variable for a
linear regression model. Consider the following multiple linear regression model:
Yi=β0+β1X1i+β2X2i+εi
where εi∼N(0, σ2)are independent and identically distributed.
Given the data set (X1i, X2i, Yi)for i= 1,2,3, . . . , n, we are interested in
testing the null hypothesis H0:β1=β2against the alternative hypothesis
Ha:β1=β2using a significance level of α= 0.05. The fitted model results in
ˆ
β1= 3.9,ˆ
β2= 4.1, and the standard error of ˆ
β1is 0.6 while the standard error
of ˆ
β2is 0.5.
Perform a hypothesis test for β1=β2using the given information.
19
Solution
Step 1: Formulate the test statistic. Given that we have to test whether β1=β2,
the test statistic is computed as
t=ˆ
β1−ˆ
β2
√SE(ˆ
β1)2+SE(ˆ
β2)2
Step 2: Compute the test statistic. Plugging in the given values, we obtain
t=3.9−4.1
√0.62+ 0.52
Step 3: Determine the critical value. Since we are performing a two-tailed
test at a significance level of α= 0.05, the critical values are ±zα/2where
zα/2=z0.025 ≈1.96.
Step 4: Make a decision. If |t|>1.96, we reject the null hypothesis H0:
β1=β2. Otherwise, we fail to reject H0.
Step 5: Calculate the test statistic value. Now, substitute the values and
compute tto determine the decision:
t=3.9−4.1
√0.62+ 0.52=−0.2
√0.36 + 0.25 ≈−0.2
√0.61 ≈ −0.255
Since | − 0.255|<1.96, we fail to reject H0.
Step 6: Formulate the conclusion. Since the test statistic value falls within
the range of the critical values, we do not have enough evidence to conclude
that β1is different from β2.
Question 22
Question
Let Y1, Y2, . . . , Ynbe independent and identically distributed random variables
following a normal distribution with mean µand variance σ2. Consider the
simple linear regression model Yi=β0+β1Xi+εi, where β0,β1, and σ2are
unknown parameters to be estimated.
Suppose that the least squares estimates of β0and β1are denoted as ˆ
β0and
ˆ
β1, respectively. Show that ˆ
β0and ˆ
β1are unbiased estimators for β0and β1,
respectively.
Solution
Step 1: Recall that the least squares estimates of β0and β1are given by:
ˆ
β1=
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
20
ˆ
β0=¯
Y−ˆ
β1¯
X
Step 2: Let’s start by showing that ˆ
β1is an unbiased estimator for β1. The
expectation of ˆ
β1can be computed as follows:
E(ˆ
β1) = E
n
∑
i=1
(Xi−¯
X)(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)E(Yi−¯
Y)
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−E(¯
Y))
n
∑
i=1
(Xi−¯
X)2
=
n
∑
i=1
(Xi−¯
X)(E(Yi)−µ)
n
∑
i=1
(Xi−¯
X)2
Step 3: Since the Yi’s are independent and identically distributed with mean
µ, we have E(Yi) = µfor all i. Thus, E(ˆ
β1) =
n
∑
i=1
(Xi−¯
X)·0
n
∑
i=1
(Xi−¯
X)2= 0.
Step 4: Therefore, ˆ
β1is an unbiased estimator for β1.
Step 5: Next, we will show that ˆ
β0is an unbiased estimator for β0. By
linearity of expectation, we have:
E(ˆ
β0) = E(¯
Y)−E(ˆ
β1¯
X)
Step 6: Since ¯
Yand ¯
Xare constants, we can simplify this expression to:
E(ˆ
β0) = ¯
Y−¯
XE(ˆ
β1)
Step 7: Since ˆ
β1is an unbiased estimator for β1, we have E(ˆ
β1) = 0. Thus,
E(ˆ
β0) = ¯
Y−0·¯
X=¯
Y.
Step 8: Therefore, ˆ
β0is an unbiased estimator for β0.
Question 23
Question
Suppose you are given the following multiple regression model:
y= 3x1−2x2+ 5x3+ 4
21
where yis the dependent variable, and x1,x2, and x3are the independent
variables. You are also given the following outputs from the regression analysis:
• Residual standard error = 2.5
•R2= 0.85
• F-statistic = 15.6
Is this model statistically significant at the α= 0.05 level? Justify your answer.
Solution
Step 1: Hypotheses
The null hypothesis H0is that all the coefficients in the regression model are
equal to zero:
H0:β1=β2=β3= 0
The alternative hypothesis H1is that at least one coefficient is not equal to
zero:
H1: At least one of β1,β2,β3is not equal to zero.
Step 2: F-statistic test
First, we calculate the degrees of freedom for the numerator and denominator
of the F-statistic:
• Numerator degrees of freedom = number of coefficients being tested = 3
• Denominator degrees of freedom = number of observations - number of
coefficients = n−k−1, where nis the number of observations and kis
the number of coefficients including the intercept term. Since we do not
have the sample size, we cannot perform the hypothesis test.
Step 3: Decision
We cannot perform the F-test without knowing the sample size n. Therefore, we
are unable to determine the statistical significance of the model at the α= 0.05
level without additional information.
Question 24
Question
Let’s consider a multiple linear regression model with three predictors X1,X2,
and X3. The estimated regression equation is given by:
ˆ
Y= 15 + 3X1−2X2+ 4X3
Assume the estimated variances for the coefficient estimates are V ar(ˆ
β0) = 10,
V ar(ˆ
β1) = 4,V ar(ˆ
β2) = 6, and V ar(ˆ
β3) = 8. Suppose we want to test the
following null hypothesis:
H0:β1= 0, H1:β1= 0
22
Perform a hypothesis test to determine whether there is a linear relationship
between X1and Yusing a significance level of 0.05.
Solution
Step 1: The test statistic for testing the hypothesis βj= 0 versus Ha:βj= 0
is given by:
t=ˆ
βj
√V ar(ˆ
βj)
Step 2: In this case, we are interested in testing the hypothesis H0:β1= 0
versus H1:β1= 0. The test statistic for β1is:
t=3
√4=3
2
Step 3: The critical value for a two-tailed test at a 5
Step 4: Using a statistical table, we find that t0.025,16 ≈2.12.
Step 5: Compare the absolute value of the test statistic to the critical value.
Since 3
2= 1.5<2.12, we do not reject the null hypothesis.
Step 6: Therefore, we do not have enough evidence to conclude that there
is a linear relationship between X1and Yat a significance level of 0.05.
Question 25
Question
Let Ybe the dependent variable and Xbe the independent variable in a
linear regression model. Suppose that the estimated regression equation is
ˆ
Y= 3 + 1.5X. The standard error of the estimate is found to be 2.5. Given
this information, test the null hypothesis H0:β1= 0 against the alternative
hypothesis Ha:β1= 0 at the 5
Solution
Step 1: Determine the test statistic using the formula:
t=ˆ
β1−0
SE(ˆ
β1)
where ˆ
β1= 1.5and SE(ˆ
β1) = 2.5. Plugging in these values, we get:
t=1.5−0
2.5= 0.6
Step 2: Find the degrees of freedom. Since we are testing the slope coeffi-
cient, the degrees of freedom is n−2, where nis the sample size. However, the
sample size is not given in the question.
23
Step 3: Find the critical value using the t-distribution table for a two-tailed
test at the 5
Step 4: Make a decision: - If |t|is greater than the critical value, reject
the null hypothesis. - If |t|is less than the critical value, fail to reject the null
hypothesis.
Step 5: State the conclusion in the context of the problem, including the
decision and the reason for it.
24
