You are on page 1of 5

Student Information

Answer 1
a)
Since the population standard deviation is unknown and the sample size is less than 30, we
should use Student’s T-distribution.

the sample size is:

n = 16
the sample mean consumption of car is:

n
X xi
= X = 6.81
i=1
n
and the sample standard deviation is :

s
Pn
− X)2
i=1 (xi
s= = 1.06
n−1
in the 98% confidence interval,

α = 0.02, α/2 = 0.01


The critical value of t distribution with n-1 degrees of freedom is tα/2 = t0.01 = 2.602 = 2.6.
Then, confidence interval for the mean consumption is:
1.06
6.81 ± (2.6) ∗ √ = 6.81 ± 0.689
16

[6.12, 7.50]

b)
we test the null hypothesis H0 : µ = 7.5 against the alternative hypothesis (one-sided left-tail)
HA : µ < 7.5. since we only interested to know if the mean gasoline consumption of car per 100
km has decreased.

The degrees of freedom = n-1 = 15 and the critical value for T-distribution :

1
t(0.05),15 = 1.753
Take t(0.05),15 as − 1.753 And rejection region for this left-tail test is R = (−∞, −1.753] (the
orange area in the graph represents the rejection region.).
at significance level α = 0.05 , from part a, we have sample statistics n = 16, X = 6.81, and s =
1.06. Compute the T-statistics,

X −µ 6.81 − 7.5
t= √ = = −2.63
s/ n 1.06/4

ta,v = −1.753 0

Our test statistics belongs to rejection region, So we can reject the null hyp. We can claim that
the improvement is effective.

c)
Since 6.5 is less than 6.81, the X − µ value will be positive. Since it is not possible for any
positive number to be in the rejection range (−∞, −1.753] , we accept the null hypothesis without
any calculation.

Answer 2
a)
we test the null hypothesis H0 : µ = 5000 against the alternative hypothesis (one-sided right-
tail) HA : µ > 5000. since we only interested to know if the mean rent price has increased. Ali’s
claim should be considered as the null hypothesis.

b)
Test statistics. we are given:

n = 100
µ0 = 5000

2
α = 0.05
σ = 2000
and from the sample,

X = 5500
The test statistics is:

X − µ0 5500 − 5000
Z= √ = = 2.5
σ/ n 2000/10
The critical value is:

za = z0.05 = 1.645
with the right tail alternative,

rejectH0 if 1.645 ≤ Z
acceptH0 if 1.645 > Z
Our test statistics Z=2.5 belongs to rejection reigon , so we can reject the null hypothesis.
since the null hypothesis was the Ali’s claim, Ahmet can claim that there is an increase in the rent
prices compared to the last year At a 5% level of significance.

c)
X−µ
P = P(Z > √0)
σ/ n

= P(Z > 2.5)

= 1-0.9938

= 0.0062
Since α = 0.05 (At a 5% level of significance) is bigger than our P-value, we reject the null
hypothesis.

d)
We know both of the standard derivation and n > 30. So, we can use Z-test.

H0 : µa = µi

3
HA : µa < µi

Xa = 5500 Xi = 6500
na = 100 ni = 60
σa = 2000 σi = 3000

α = 0.01
z0.01 = −2.33

(Xa − Xi ) − (µa − µi )
Z= q
σa2 σ2
na
+ nii
5500 − 6500
=q
20002 2
100
+ 3000
60

−1000
=
435.89

= −2.294
−2.294 > −2.33. So, our test value remains in the accepting region. This implies that when
the data does not provide enough evidence to reject it, we accept the null hypothesis as the default
option.

Answer 3
H0 : The number of rainy days in Ankara is independent of the season.
HA : The number of rainy days in Ankara is not independent of the season.

100∗90 100∗90
Êxp(1, 1) = 360
= 25, Êxp(1, 2) = 360
= 25,

100∗90 100∗90
Êxp(1, 3) = 360
= 25, Êxp(1, 4) = 360
= 25

260∗90 260∗90
Êxp(2, 1) = 360
= 65, Êxp(2, 2) = 360
= 65,

260∗90 260∗90
Êxp(2, 3) = 360
= 65, Êxp(2, 4) = 360
= 65
(34−25)2 (32−25)2 (15−25)2 (19−25)2
x2obs = 25
+ 25
+ 25
+ 25
+
(56−65)2 (58−65)2 756−65)2 (71−65)2
65
+ 65 + 65 + 65
= 14.73

4
P = P {x2 ≥ 14.73}
Based on the information provided in Table A6 with 3 degrees of freedom, the obtained p-value
falls within the range of [0.001, 0.005]. Since the p-value is less than 0.01, we can conclude that
there is enough evidence to reject the null hypothesis H0 . Therefore, it can be inferred that the
number of rainy days in Ankara is not dependent on the season.

Answer 4
X = [34 32 15 19; 56 58 75 71];
Row = sum(X, 2);
Column = sum(X, 1);
Sum = sum(Row);
e = zeros(size(X));
for i=1:length(Row)
for j=1:length(Column)
e(i,j) = Row(i)*Column(j)/Tot;
end
end
chisquare = (X-e).^2./e;
chix = sum(sum(chisquare));
P_value = 1-chi2cdf(chix,(length(Column)-1)*(length(Row)-1))
X_obs = chix

You might also like