UNIVERSITY OF THE WEST INDIES

CAVE HILL CAMPUS

ECON 1005 – Introductory Statistics
Tutorial 9 – Correlation & Regression

1. Dr. Alleyne collected data on the number of absences from tutorials (X) and a student’s
score on the Stats final exam (Y). A sample of 6 students yielded the following
information:

Number of absences (X) Score (Y)

1 95
3 86
6 70
9 60
4 20
2 92

a. Calculate the correlation coefficient.

b. Develop the estimated simple regression equation, that is, provide the values of
a and b for the equation 𝑌" = a + bX
c. Interpret the intercept and the slope of the regression.
d. Predict a student’s score in stats if they were absent for 7 tutorials
e. Given your answer in part (a), what is the value of the coefficient of
determination (that is r2) for the regression.
f. Interpret the r2 calculated in part (e).

𝑥̅ = 4.167 𝑦& = 70.5

X Y X – 𝑥̅ Y – 𝑦& (X – 𝑥̅ )2 (Y – 𝑦&)2 (X – 𝑥̅ ) (Y – 𝑦&)

1 95 -3.167 24.5 10.030 600.25 -77.592
3 86 -1.167 15.5 1.362 240.25 -18.089
6 70 1.833 -0.5 3.360 0.25 -0.9165
9 60 4.833 -10.5 23.358 110.25 -50.7465
4 20 -0.167 -50.5 0.028 2550.25 8.4335
2 92 -2.167 21.5 4.696 462.25 -46.5905
42.834 3963.5 -185.501

!(# – &̅ ) () – *+) -./0.02.

Cov (x, y) = ,-.
= 3-.
= -37.1002

!(# – &̅ )4 54./65

S2x = = = √8.5668 = 2.927
,-. 3-.

!(7 – *+)4 6836.0

S2y = ,-.
= 3-.
= √792.7 = 28.15

9:;(#,)) -6?..224
a. r = =4> =47
= (4.84?)(4/..0)
= −0.45

!(# – &̅ ) () – *+) -./0.02.

b. b = !(# – &̅ )4
= 54./65
= -4.33

𝑎 = 𝑦& - b𝑥̅ = 70.5 – (-4.33x4.167) 88.55

 𝑦6 = 88.55 – 4.33X

c. For a person with no absences, their score is predicted to be 88.55.

For every additional absence, the score will decrease by 4.33.

d. x = 7 𝑦6 = 88.55 + (-4.33)(7) = 58.24

e. r2 = (-0.45)2 = 0.203

f. 20% of the variation the score is explained by the variation in number of

absences.

2. You are given the following information about y and x.

Dependent Variable (y) Independent Variable (x)

5 1
4 2
3 3
2 4
1 5

Find:

a. The least squares estimate of the slope or b1.

b. The least squares estimate of the intercept or b0.
c. The point estimate of y when x = 2.
d. The sample correlation coefficient.

𝑥̅ = 3 𝑦& = 3

Y X Y – 𝑦& X – 𝑥̅ (Y – 𝑦&)2 (X – 𝑥̅ )2 (X – 𝑥̅ ) (Y – 𝑦&)

5 1 2 -2 4 4 -4
4 2 1 -1 1 1 -1
3 3 0 0 0 0 0
2 4 -1 1 1 1 -1
1 5 -2 2 4 4 -4
10 10 -10

!(# – &̅ ) () – *+) -.2

a. b1 = !(# – &̅ )4
= .2
= -1

b. b0 = 𝑦& - b𝑥̅ = 3– (-1x3) 6

c. 𝑦6 = 6 + (-1x2) = 4

!(# – &̅ ) () – *+) -.2

d. Cov (x, y) = ,-.
= 0-.
= -2.5

!(# – &̅ )4 .2
S2x = ,-.
= 0-.
= √2.5 = 1.58

!(7 – *+)4 .2
S2y = ,-.
= 0-.
= √2.5 = 1.58
9:;(#,)) -4.0
r = =4> =47
= (..0/)(..0/)
= -1.00
 3. A recent study was carried out to evaluate the relationship between the amount of
alcohol consumed (ounces) by students and the number of social events attended during
the semester. After collecting information on the average alcohol consumption of 100
students and the number of social events they attended semester, a researcher
estimated the following simple regression

y = 𝛼 + 𝛽𝑥 + ei

which yielded the following estimates

𝑦6= 10 + 16x

where y is the average amount of alcohol consumed by students and x represents the number of
social events attended during the semester.

a. Which of the following scatterplots (not drawn to scale) best describes the
relationship between alcohol consumption and the number of social events
students attended during the semester? Give a reason for your answer

b Since the relationship is positive.

b. Interpret the intercept in this equation.

For every additional social event attended, the alcohol consumption increases by
16 ounces
c. Interpret of the slope in this equation.
If no social events are attended, the alcohol consumption is expected to be 10
ounces.

4. For the following data, the value of SSE = 18.

Dependent Variable (y) Independent Variable (x)

15 4
17 6
23 2
17 4
 Find:

a. The slope of the regression equation.

b. The y-intercept.

𝑥̅ = 4 𝑦& = 18

Y X Y – 𝑦& X – 𝑥̅ (Y – 𝑦&)2 (X – 𝑥̅ )2 (X – 𝑥̅ ) (Y – 𝑦&) 𝑦6 𝑦6 – 𝑦& (𝑦6 – 𝑦&) 2

15 4 -3 0 9 0 0 18 0 0
17 6 -1 2 1 4 -2 15 -3 9
23 2 5 -2 25 4 -10 21 3 9
17 4 -1 0 1 0 0 18 0 0
36 8 -12 18

!(# – &̅ ) () – *+) -.4

b. b = = = -1.5
!(# – &̅ )4 /

𝑎 = 𝑦& - b𝑥̅ = 18 – (-1.5)(4) = 24

5. The following information regarding a dependent variable y and an independent

variable x is provided:

Σx=90 Σ(y - 𝑦&)(x - 𝑥: )= - 156

Σy=340 Σ(x - 𝑥: )2 = 234
n=4 Σ(y - 𝑦& )2=1974
SSR = 104

Find:

a. The total sum of squares (SST).

b. The sum of squares due to error (SSE).
c. The slope of the regression equation.
d. The y-intercept.
e. The coefficient of correlation.

a. SST = Σ(y - 𝑦& )2 = 1974

b. SSE = SST – SSR = 1974 – 104 = 1870

!(# – &̅ ) () – *+) -.03

c. b = !(# – &̅ )4
= 465
= -0.667

d. 𝑎 = 𝑦& - b𝑥̅ = 85 – (-0.667)(22.5) = 85 + 15.0075 = 100.0075

 !(# – &̅ ) () – *+) -.03
e. Cov (x, y) = ,-.
= 5-.
= -52

!(# – &̅ )4 465

S2x = ,-.
= 5-.
= √78 = 8.83

!(7 – *+)4 .8?5

S2y = = = √658 = 25.65
,-. 5-.

9:;(#,)) -04
r = = = -0.23
=4> =47 (/./6)(40.30)