Você está na página 1de 10

GROUP ASSIGNMENT 2

Statistics I International business

Group: 80
Youngjun Mo: 2825732

Tutorial group 11-12

Nikolay Aleksiev: 2845423

Tutorial group 11-12

Josh O'Mahony Westerhof: 2913712

Tutorial group 15-16

Question 1
a) Estimate the least square line:
We use the following formula: ^ = b0 + b1 xi
is the depended variable, ^
slope and b0

where x is the independed variable and y

is the estimated value of the depended variable b1 is the

the intercept.

From the table above we derive that the least square line is: ^ = 12.038 + 0.0674 x i

Least squares line


12.21

12.27
12.18

12.03

f(x) = 0.07x + 12.04 11.82


R = 0.74

11.92

11.83
Time of 100m sprint

11.7

11.42

-10.00

11.57

-8.00

-6.00

-4.00

-2.00

0.00

2.00

Wind in Km/h

b) Interpret b1 = 0.067430523 x i
Interpretation: For every km/h of headwind it is expected that the running time of Dafne
will increase on average with 0.0674 seconds. The reverse is in power for tailwind.
2
Calculate the coefficient of determination R :

Refer to the Table in 1a)

R2=

0.54
0.72 = 0.7445

Interpretation: 74% of the variation in the sprint-time achieved by Dafne is explained by


the variation in the headwind/tailwind speed.

c) Output of excel using Data analysis tool for the regression model:

^ = 12.038 +

0.0674 x i at 95% significance level. P-value < =0.05 hence we reject


H 0 : 1=0 and accept

H 1: 1 0

(P-value=0.001307742)

From Excel we derive the following


information on the left:

Test the hypothesis weather the wind indeed has an effect on Dafnes 100m times:
1) n = 10
2)

b1 = 0.067 S b = 0.014 Significance level 5% => =0.05

H 0 : 1=0 versus

= 0.025

H 1: 1 0

3) The test statistics is


=

0.067
0.014

4.828
4) If H 0 is true then the test-statistic is t-distributed with 8 degrees of freedom.
5) Reject H 0 if t tn2, = t 8,.025 = 2.306 and t t n2, = t 8,. 025 = -2.306
p-value = P(t>4.828)*2 =~0.001*2= ~0.002 (0.001307742 exactly)
p-value (0; 0.002) interval
6) We reject H 0 because 4.828>2.306 and accept H 1

7) There is sufficient evidence to claim that the wind indeed has an effect on Dafnes
performance. Tailwind helps her achieve better times and headwind slows her down.

Question 2
Jack breeds rabbits. He is searching for new female Flemish Giants. He considers two fellow
breeders to buy these females from. Jack is interested in the amount of rabbits in one nest. He
wants to buy rabbits from the breeder with the largest nests. He asks data from both breeders
to make a comparison. The first breeder, Shirley, sends data about her last 14 nests, the
second breeder, Simona sends data about her last 12 nests. Jack computes the following
sample statistics using this data:
Shirleys nests have an average size of 6.9 rabbits, with a standard deviation of 1,8.
Simonas nests have an average size of 9.8 rabbits, with a standard deviation of 3.7.
Assume normally distributed nest sizes.
Question 2a

Test whether there is difference in the population variances of the nest sizes for the breeders
at 10% significance level.
In order to derive the answer of this question we will test the following hypothesis:
1) Shirley: n1 =14

x 1 =6.9

S 1 =1.8

S 21 =3.24

Simona: n2 =12

x 2 =9.8

S 2 =3.7

S 22 =13.69

H0;

21
22 =1 against

H1;

21
1
22

2) We test at 10% significance level => =0.1 and

=0.05

3) Test statistic is:


2

s1
F= s 2
2

3.24
= 13.69 0.24

H 0 is true the test-statistic if F-distributed with v 1 = n1 -1 and v 2 = n2 -1

4) If
=>

(14-1)+(12-1) = 24 degrees of freedom.


5) Reject
or if

H 0 if F > Fa / 2,v 1,v 2 = F05,13,11 = 2.76


F < F1a /2, v 1,v 2 = F95,13,11 = 0.38

6) Since F =0.24 <

F95,13,11 = 0.38 we reject

H 0 and accept the alternative hypothesis

H1.

7) Our test-statistic is in the rejection region therefore we have sufficient evidence not to
2
2
claim equal population variance. We are 90% sure that 1 2 .
Question 2b
Test whether there is a difference in the mean nest sizes of breeders at 10% significance level.

Referring to Question 2a, we assume unequal population variances 1 2 .


2

1) Shirley: n1 =14

x 1 =6.9

S 1 =1.8

S 1 =3.24

Simona: n2 =12

x 2 =9.8

S 2 =3.7

S 22 =13.69

H 0 ; 12= 0
H 1 ; 1 2 0

2) We test at 10% significance level => =0.1 and

=0.05

3) Test statistic:
t=

x 1 x 2

6.99.8


2
1

2
2

S
S =
+
n1
n2

4) When

or if

2.9
1.37

2.48

H 0 is true our test-statistic is student-t distributed with 15 degrees of freedom:

S 21
3.24
n1
14

=
2
2

v =

5) Reject

3.24
13.69
+
14
12

=> v =15

H 0 if t> t v , = t 15,.05 =1.753


2
t<t v , /2 = t 15,.05=1.753

7) We reject

H 0 because t=2.48<t 15,.05=1.753 and accept the alternative

hypothesis H 1

7) There is sufficient evidence to claim that there is difference in the population means of the
nest size of breeders.

Question 2c
Simona is a close competitor of Jack. He decides that he only buys rabbits from Simona if her
nests, on average, have two more rabbits than Shirleys nests. Will Jack buy Simonas rabbits
(use a 5% significance level)?

3) Shirley: n1 =14

x 1 =6.9

S 1 =1.8

S 21 =3.24

Simona: n2 =12

x 2 =9.8

S 2 =3.7

S 22 =13.69

H 0 ; 21=2
H 1 ; 2 1> 2
4) We test at 5% significance level => =0.05
3) Test statistic:
t=

x 2 x 1(21 )


2
1

2
2

S
S
+
n1
n2

4) When

3.24
13.69
+
14
12

0.9
= 1.17

0.77

H 0 is true our test-statistic is student-t distributed with 15 degrees of freedom:

S 21
3.24
n1
14

=
2
2

v =

5) Reject

( 9.86.9)2

=> v =15

H 0 if t> t v , = t 15,.05 =1.753

6) We fail to reject

H 0 because t=0.77<t 15,.05=1.753

and therefore do not accept the

alternative hypothesis H 1
7) There is insufficient evidence to claim that Simonas nests have on average more than two
rabbits compared to Sherleys. However, Jack is willing to buy from Simona if she has on
average two more not more than two so, he might as well do business with her. Another
alternative that would bring more evidence whether Jack should buy only from Simona is to
test the following hypothesis:
H 0 ; 21=1

I.

H 1 ; 2 1>1
In that case we are testing whether she has more than one rabbits. If we fail to reject

H0

this would mean that Simona doesnt have more than 1 rabbits which ultimately leads to the
conclusion that Jack wont buy from her.
II.

In this scenario our test statistic is:


t=

III.

x 2 x 1(21 )


2
1

2
2

S
S
+
n1
n2

When

( 9.86.9)1

3.24
13.69
+
14
12

1.9
= 1.17

1.62

H 0 is true our test-statistic is student-t distributed with 15 degrees of freedom:

S1
3.24
n1
14

=
2
2

v =

IV.

V.

Reject

=> v =15

H 0 if t> t v , = t 15,.05 =1.753

We fail to reject

H 0 because t=1.62<t 15,.05 =1.753 and therefore do not accept the

alternative hypothesis H 1

VI.

There is insufficient evidence to claim that Simona has more than 1 rabbit in her nests
compared to Shirley which ultimately means that she doesnt have 2 either, as a result
Jack wont buy from Simona.

Você também pode gostar