STAT3302 Final Exam Jan.

12, 2022 Time: 2 hours

Name:
ID Number:

Question: 1 2 3 4 5 6 Total
Points: 10 6 6 6 6 6 40
Score:

1. (10 points) Circle True or False to indicate the validity of the following statements. You do not need to
justify your answer. In all of the following, assume that X1 , X2 , . . . , Xn is a random sample from some
population of interest with mean µ and variance σ 2 .
(a) TRUE / FALSE Var(X1 + 2X2 ) = Var(X1 ) + 2Var(X2 ) when the X1 and X2 are independent.
(b) TRUE / FALSE The sum of χ2 random variables has a χ2 distribution when the random
variables are independent.
(c) TRUE / FALSE For the normal distribution N (µ, σ 2 ), X is the minimum-variance unbiased
estimator for µ.
Pn
(d) TRUE / FALSE n1 i=1 (Xi − X)2 is an unbiased estimator for σ 2 .
(e) TRUE / FALSE A t statistic with n degrees of freedom is defined as t = √ZU where Z ∼ N (0, 1)
and U ∼ χ2 (n).
Pn 3
(f) TRUE / FALSE The method of moments estimator for E(X 3 ) is n1 ( i=1 Xi ) .
(g) TRUE / FALSE You measured the temperature on 5 randomly chosen days in January, 2011
in Gaza. Assuming the data are normal, a 95% CI for the true mean January
temperature is X ± 1.96 √Sn .
(h) TRUE / FALSE A 95% confidence interval forµ1 − µ2 is [−6, −1]. Thus, with 95% confidence,
µ2 > µ1 .
(i) TRUE / FALSE If F has an F distribution with m and n df, then 1/F has an F distribution
with n and m df.
(j) TRUE / FALSE Var(χ2n ) = n.
2. Suppose that Y1 , Y2 is a random sample from N (µ, σ 2 ).
(Y1 −Y2 )2
(a) (3 points) What is the distribution of 2σ 2 , and explain why your answer is correct.

(Y1 −Y2 )2
(b) (3 points) Use 2σ 2 to construct a 100(1 − α)% confidence interval for σ 2
3. (a) (2 points) Use the sample Y1 = 3.9, Y2 = 0.8, Y3 = 4.3, Y4 = 4, Y5 = 1.3, Y6 = 1.7 and Y7 = 7.1 to
calculate the maximum likelihood estimate for λ in the exponential pdf

fY (y; λ) = λe−λy , y ≥ 0.

(b) (2 points) Let Y1 , Y2 , . . . , Yn be a random sample of size n from the pdf

(c) (2 points) Suppose that H0 : µ = 25.0 is to be tested against H1 : µ > 25.0 using a sample of size
n = 30 and a standard deviation σ = 2.4. The decision rule states that H0 should be rejected if
y ≥ 25.718. Compute the probability of committing a Type I error.

Page 2
4. Let Y1 , Y2 , . . . , Yn be a random sample of size n from
1 −1y
fY (y; θ) = e θ , y > 0.
θ

(a) (2 points) Show that θ̂1 = Y1 , and θ̂2 = Y are unbiased estimators for θ.

(b) (2 points) Calculate the relative efficiency of θ̂2 to θ̂1 .

(c) (2 points) Let X1 , X2 , · · · , Xn be a random sample of size n from the Poisson distribution,

e−λ λk
pX (k; λ) = , k = 0, 1, · · · .
k!
Find an efficient estimator for λ. Prove your claim.

Page 3
5. To test H0 : p = 0.40 versus H1 : p > 0.4, where p is the probability of success of some event, n = 52
independent Bernoulli trials were conducted. If these trials result in k = 24 successes.
(a) (3 points) perform the test at the α = 0.05 level of significance.

(b) (3 points) Calculate the p-value for these data. For what values of α would H0 be rejected?

Page 4
6. (a) (3 points) Two independent data drawn from two normal distributions with common unknown
variance σ 2 and means µX and µY , respectively are summarized in the following table.
x y
n = 119 m = 134
x = 3.2 y = 3.5
sX = 0.16 sY = 0.18
Test the H0 : µX = µY versus H1 : µX < µY at the α = 0.05 level of significance.

(b) (3 points) Consider the two sets of independent data in the following table.
x y
2.43 0.81
1.83 0.70
2.43 0.74
2.70 0.36
1.88 0.75
1.96 0.56
1.53 0.65
2.08 0.87
1.85 0.40
2.44 0.31
x = 2.11 y = 0.62
sX = 0.37 sY = 0.20
Construct a 95% confidence interval for σX /σY and test the null hypothesis H0 : σX = σY versus
H1 : σX 6= σY at the α = 0.05 level of significance.

Page 5