Question 1.
Consider a stratified design composed of H strata of size Nh, h =
1,...,H. We want to estimate the population mean µy of the
characteristic y. Let µx,h, h =1,...,H be the means in the strata
(in the population) of an auxiliary characteristic x. The µx,h are
supposedly known and we propose to estimate µy using the following
estimator: b µD = yst +µx −xst where yst and xst are the basic
estimate of the population means µy and µx for y and x,
respectively.
(a) Give an expression of µx in terms of µx,h, h =1,...,H.
(b) Show thatb µD is unbiased estimates for µy.
(c) Give the variance ofb µD
(d)Let n =PH h=1 nh be the sample size. What is the optimal allocation of the nh in order to minimise the variance ofb µD? We consider that the init cost of the survey does not depend on the stratum. (e) (1 mark) In which favourable case isb µD preferable to yst?
Question 1. Consider a stratified design composed of H strata of size Nh, h = 1,...,H....
photos for each question are all in a row
(1 point) In the following questions, use the normal distribution to find a confidence interval for a difference in proportions pu - P2 given the relevant sample results. Give the best point estimate for p. - P2, the margin of error, and the confidence interval. Assume the results come from random samples. Give your answers to 4 decimal places. 300. Use 1. A 80% interval for pı - P2 given that...