F T ra n sf o F T ra n sf o

PD rm PD rm
Y Y
Y

Y
er

er
ABB

ABB
y

y
bu

bu
2.0

2.0
to

to
re

re
he

he
k

k
lic

lic
C

C
w om w om
w

w
w. w.
A B B Y Y.c A B B Y Y.c

Test Exercice 1

a) Diagram of Sales and Advertising

50
40
Sales
30
20

6 8 10 12 14 16
Advertising

I would find a line regression with a positive slope. And find how increased advertising will increase
sales.

b)

sales Coef. Std. Err. t P>t [95% Conf. Interval]

The coefficient and are 29.62 and -.32 respectively. b is no significantly different from 0.

c)
 F T ra n sf o F T ra n sf o
PD rm PD rm
Y Y
Y

Y
er

er
ABB

ABB
y

y
bu

bu
2.0

2.0
to

to
re

re
he

he
k

k
lic

lic
C

C
w om w om
w

w
w. w.
A B B Y Y.c A B B Y Y.c

.15
.1
Density
.05
0

-5 0 5 10 15 20
Residuals

The distribution of the residuals is not normally distributed

d) To get a satisfactory regression model we could only consider sales of the day or drop the
observation.

e)

Sales Coef. Std. Err. t P>t [95% Conf. Interval]

advertising 0.375 .0881964 4.25 0.001 0.1889218 0.5610782

The coefficient and are 21.125 and .375 respectively. . b is significantly different from 0

f) the differences between the coefficients are that in the observqacion was removed and the largest
resiudal. A high residual in observation of the week 12 shows that there are other factors that increased
sales. And if we drop the observation 12, we get the expected regression line.