MAT 4179-5192 Assignment 3 Fall 2023
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Assignment 3
MAT 4179-5192
Fall 2023
To hand-in on
October 4, 2023 on Brightspace before midnight
(pdf format only)
Problem 1
Exercise no. 1 page 66 in Lohr (2022, Chapter 2, Section 2.13).
Problem 2
Show that Expression (4.9) in the course notes reduces to Expression (4.10) in the case
of simple random sampling without replacement.
Hint: Don’t forget to separate the case i = j from the case i j.
Problem 3
Consider a population U of size N. From U, we select a sample, S, of size n, according to a sampling design with first-order inclusion probabilities πi and second-order inclusion probabilities πij , i j. For which finite population parameter is the following estimator unbiased?
Simplify your answer and you will obtain a familiar parameter.
Problem 4
Show that the design effect of Bernoulli sampling is given by
where CV(y) = Sy /yU denotes the coefficient of variation of the survey variable y.
Problem 5
(FOR GRADUATE STUDENTS ONLY) Consider a population U of size N. We consider a sampling design that leads to N + 1 samples: N samples are of size 1, where each sample contains one of the possible N elements; one sample is of size N, containing all the population elements. Each of the N + 1 possibles is given an equal probability of
selection. Note that this is a random-size sampling design.
(a) Determine the first-order and the second-order inclusion probabilities for this sam-pling design.
(b) Determine the expected sample size.
(c) Determine the variance of the sample size.
(d) As an estimator of the population total, ty =Σk=U yk , we consider the following estimator:
where ns denotes the realized sample size. Show that ty is unbiased for ty .
Hint: For (d), you may want to use the law of total expectations.
2023-10-21