Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 4

Jerome D.

Magalad BSE 3-1 Economic Statistics

Answers:

4.3

A.If seven people is taken out of a sample of 50 were left-handed.The true proportion would be 14%
(7/50).

B.If 90% confidence is being compared with 95% the result would be 1.64 instead of 1.96, giving 0.14 ±
0.080.

C.By using the 99% to be compared to others would be 0.14 ± 0.126.

4.4.

X1 − X2 = 25 − 30 = −5 is the point estimate. The interval estimate is given by

4.5

A survey of 50 16-year old girls revealed that 40% had a boyfriend. A survey of 100 16-year old boys
revealed 20% with a girlfriend. The point estimate is 20%. The interval estimate is

4.6

A sample of size n = 16 is drawn from a population which is known to be Normally distributed. The
sample mean and variance are calculated as 74 and 121.When it was founded and calculated by
99%.The 99% CI is given by 74 ± t* ×√121/16 = 74 ± 2.947 × 2.75 = 74 ± 8.10 =[65.90, 82.10].

4.7

The 95% confidence interval estimate of the difference between the two populationmeans that

The 95% CI is therefore

(45-55)+/- 2.042x √391.36/15 + 391.36/20 =− 10± 13.80=[-3.8 23.8]


5.1

A.Two political parties are debating crime figures. One party says that crime hasincreased compared to
the previous year. The other party says it has not.The crime rate has the same spot as last year.At H1,it
has increased.

B.The Type I Error when the situation affects the assumption that crime rate has risen,but in true fact,it
hasn't increase at all.While in Type II Error,it assumed that the crime hasn't increase but the true fact
was it had increased.

5.2

A. If the mean is greater than 108. If in fact e ~ N(100, 900/25),the probability ofmaking a Type I error is
z = (108 − 100)/√36 = 1.33.

B.If we wanted a 5% Type I error probability the result we will adopt would be z = 1.64 cuts off 5% in the
upper tail of the distribution, hence we need thedecision rule to be at X + 1.64 × s/√n = 100 + 1.64 × √36
= 109.84.

C.If we knew that μ could only take on the values 100 (under H0) or 112 (under H1) would be the Type II
error probability using the decision rule in part A is under H1: μ = 112, we can write X ~ N(112,
900/25).Hence z = (108 − 112)/√36 = −0.67.This gives an area in the tail of 25.14%, which is the Type II
error probability.

5.3

If Hypothesis of H0: μ = 500 versus H1: μ ≠ 500 using the evidence e = 530, s = 90 from a sample of size n
= 30 therefore α = 0.05 the critical value is z* = 1.96. The test statistic is z = (530 − 500)/(90/√30) = 1.83 <
1.96 so H0 is not rejected at the 5% significance level.So,that is the hypothesis of it.

5.4

If a researcher believes the cost of making a Type I error is much greater than the cost of a Type II error,
they should choose 1% significance level because hence set a low significance level (1%) so that H0 is
only rejected by very strong evidence.

5.5

A. The program reports a test statistic of z = 1.77 (P value = 0.077).(i) it would be REJECT. The Prob-value
should be halved, to 0.0385, which is less than 5%. Alternatively, 1.77 > 1.64.(ii) DO NOT REJECT, the
Prob-value is greater than 5%; equivalently 1.77 < 1.96.So,that would be the result.

B.In using a 1% significance level, in the one-tailed case, 0.0385 > 1%, so the null is not rejected.
5.6

A survey of 80 voters finds that 65% are in favour of a particular policy and that 50% would prove the
hypothesis is true,

𝟎.𝟔𝟓−𝟎.𝟓
=2.68
𝟎.𝟓−𝟎.𝟓

𝟖𝟎

So the null is absolutely rejected.

5.7

A survey of 50 teenage girls found that on average they spent 3.6 hours per week chatting with friends
over the internet. The standard deviation was 1.2 hours. A similar survey of 90 teenage boys found an
average of 3.9 hours, with standard deviation 2.1 hours.Their behaviour between them would be

We have the data: X1 = 3.6, s1 = 1.2, n1 = 50; X2 = 3.9, s2 = 2.1, n2 = 90. The null hypothesis is H0: μ1 =
μ2 versus H1: μ1 ≠ μ2. The test statistic is

5.8

One gambler on horse racing won on 23 of his 75 bets. Another won on 34 out of 95.The solution will
goes like this:

The Pooled variance is,

And the test statistic is,

In conclusion, This is less in absolute magnitude than 1.64, the critical value of a one tailed test, so the
null is not rejected. The second gambler is just luckier than the first.

5.9

It is asserted that parents spend, on average, £540 per annum, on toys for each child. A survey of 24
parents finds expenditure of £490, with standard deviation £150.Well, The hypothesis is H0: μ = 540 vs
H1: μ ≠ 540. Given the sample evidence, the test statistic is

t distribution for 23 degrees of freedom is 2.069, so the null is not rejected.


5.10

The hypothesis to test is H0: μF − μN = 0 vs H1: μF − μN > 0 (F indexes finalists, N the new students). The
pooled variance is calculated as

The null is decisively rejected and we conclude finalists do spend more time in the library.

5.11

By the method of independent samples we obtain X1 = 13, X2 = 14.5, s1 = 4.29, s2 = 3.12, with n = 12 in
both cases. The test statistic is therefore

You might also like