1) The document provides solutions to questions from Statistics 332 Assignment 1.
2) It includes calculations for estimating population proportions from sample data, the variance of the sample mean for a simple random sample without replacement, and the variance of a sample proportion.
3) It also discusses four possible systematic samples that could be drawn and calculates their variance, showing it is smaller than the simple random sample variance due to the ordered nature of the population values.
1) The document provides solutions to questions from Statistics 332 Assignment 1.
2) It includes calculations for estimating population proportions from sample data, the variance of the sample mean for a simple random sample without replacement, and the variance of a sample proportion.
3) It also discusses four possible systematic samples that could be drawn and calculates their variance, showing it is smaller than the simple random sample variance due to the ordered nature of the population values.
1) The document provides solutions to questions from Statistics 332 Assignment 1.
2) It includes calculations for estimating population proportions from sample data, the variance of the sample mean for a simple random sample without replacement, and the variance of a sample proportion.
3) It also discusses four possible systematic samples that could be drawn and calculates their variance, showing it is smaller than the simple random sample variance due to the ordered nature of the population values.
1) The document provides solutions to questions from Statistics 332 Assignment 1.
2) It includes calculations for estimating population proportions from sample data, the variance of the sample mean for a simple random sample without replacement, and the variance of a sample proportion.
3) It also discusses four possible systematic samples that could be drawn and calculates their variance, showing it is smaller than the simple random sample variance due to the ordered nature of the population values.
Copyright:
Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online from Scribd
Download as pdf or txt
You are on page 1/ 2
Stat 332 Solutions to Assignment 1
1. Each question worth 2pt, except #4 worth 4pt. (a) 1) All likely voters reside in Kitchener-Waterloo area. 2) A clustered population, with each cluster a household; 3) The sampling frame consists of a list of household telephone numbers in KitchenerWaterloo area. 4) The sampled population consists of all residents live in the households that on the telephone lists and who could be reached during the phone call and are willing to answer the question. 5) yk = 1 if the k th resident says he/she will vote for Liberal candidate, yk = 0 otherwise. 6) The proportion of likely voter who will vote for the Liberal candidate. (b) 1) All Canadians aged 22 or older 2) A stratied population with each strata a province. 3) The sampling frame can be viewed as a simple stratied lists of residential phone numbers for provinces if each phone number represents a unique person; it is a stratied list of clusters if each phone number represents a house (a group of people). 4) All Canadians who live in one of the provinces, who can be reached by phone, and who is willing to answer the survey question. 5) Since each element (person) belongs to one and only one of the three groups, we can use two indicator variables: yij = 1 if the j th person in province i answers A, yij = 0 otherwise; zij = 1 if the j th person in province i answers D, zij = 0 otherwise; and yij = 0 and zij = 0 represent the answer U . 6) The three population proportions who answers A, D or U among all Canadians under , P2 = Z and P3 = 1 P1 P2 . study, i.e. P1 = Y (Two alternative ways would be (i) using a single y taking values 1, 2 and 3 to represent the three answers or (ii) using three indicator variables to represent the three answers) 2. (5pt) ( )
)] = E E [V (t
kS S
k, k k, (
yk y k ) )
= E
Ik I (
kU U
k, k k,
yk y k yk y k
k, k = E [Ik I ] k, kU U ( )
k, k yk y = k, k, k kU U yk y = (k, k ) k kU U ) = V (t 3. (a) (4pt) y U = 86.5; S 2 = 245.
2 3. (b) (4pt) The value y S and SS depend on the actual sample you obtained. (4pt) E ( yS ) = y U = 86.5; V ( yS ) = (1 1/4)S 2 /5 = 36.75. 2 (2pt) V ( yS ) is computed by plug-in SS .
3. (c) (2pt)P = 5/20 = 1/4 (4pt) get p s is dened as s = m/n from your data; the standard error of an estimator p s.e.( ps ) = V ( ps ), where V ( ps ) = (1 1/4)[(1/4) ps (1 p s )] is the estimated variance and you need to plug in the value of p s from the sample you have drawn, i.e. p s = m/5, where m is the number of students in your sample who are overweight. 3. (d) (2pt) The four possible samples are s1 = {1, 5, 9, 13, 17} (with yi s being {60, 75, 85, 90, 100}), s2 = {2, 6, 10, 14, 18} (with yi s being {60, 80, 85, 95, 105}), s3 = {3, 7, 11, 15, 19} (with yi s being {65, 80, 90, 95, 110}), and s4 = {4, 8, 12, 16, 20} (with yi s being {70, 80, 90, 100, 115}). The four sample means are 82, 85, 88 and 91. (3pt) V ( yS ) = [(82 86.5)2 + (85 86.5)2 + (88 86.5)2 + (91 86.5)2 ]/4 = 11.25 (2pt) This is much smaller than the variance from (b). This is due to the fact that the population values of yi s are in a clear increasing order and systematic sampling will do better than SRSWOR in this case.
[FREE PDF sample] Clinical Statistics Introducing Clinical Trials Survival Analysis and Longitudinal Data Analysis Jones and Bartlett Series in Mathematics 1st Edition Olga Korosteleva ebooks
