Lexi Zhang
Assignment HW02 due 10/09/2015 at 06:00pm PDT
STAT344-101 2015W1
The le lawyers.csv linked below contains the result of a
simple random sample of 35 out of 220 counties that comprise
the mythical land of Nod. Each row corresponds to a sampled c
Lexi Zhang
Assignment HW01 due 09/25/2015 at 06:00pm PDT
STAT344-101 2015W1
Your market research rm has been hired to conduct a survey. The goal is to produce estimates of both average annual
income and proportion with a post-secondary credential (PSC)
in
STATISTICS 344, Lecture #10
Estimation in Domains
October 2, 2015
Clicker Q
Think of a pair of variables (X , Y ) in a population of size N, from
which a simple random sample of size n will be drawn. Considering
our usual notation for sample and populatio
Statistics 344 - Final Exam
Wednesday April 13, 2011
Instructions: Two sheets (both sides) of notes and a simple calculator are allowed. The
exam time is 150 minutes. There are eight problems, with a total of 80 points available.
This booklet has 11 pages
STATISTICS 344, Lecture #1
Some First Thoughts on Sample Surveys
September 9, 2015
Basic elements
Characteristic(s) of interest - what we measure.
Observation unit - who/what we measure it on.
characteristic
income
height
# of people
observation unit
pers
STATISTICS 344, Lecture #3
The Combinatorial Approach to Random
Sampling
September 14, 2015
Is Independence exactly right?
A fairly standard intro stat textbook denition of random sampling
says that:
Everyone has the same chance of being included.
Whether
STATISTICS 344, Lecture #5
Study Planning
September 18, 2015
Clicker Q: Put yourself in the mindset of a research
funding agency.
. Many potential scientic studies are pitched to you, but there
are only resources to fund some of them. Amongst proposed
stu
STATISTICS 344, Lecture #2
Some Review, and a Demo
September 11, 2015
STAT 200: Inference for/by a Mean
With respect to a numerical variable Y , typical notation would be:
Size
Mean
SD
Population
N
(ANIYB)
p
p(1 p)
Sample
n
y
s
ANIYB = Alternate Notation
STATISTICS 344, Lecture #9
Regression Estimation
September 30, 2015
Recall from last time: A graphical view of ratio estimation
Think of line passing through (0, 0) and (S , yS ) as a tted line
x
or trend line.
Ratio estimation corresponds to using this
STATISTICS 344, Lecture #4
Tweaking the STAT 200 Theory
September 16, 2015
Scorecard (i): Population mean, total, variance:
yP
=
1
N
N
yi
i=1
N
tP
=
yi
i=1
2
sP
=
1
N 1
N
(yi yP )2 .
i=1
Scorecard (ii): Sample mean, total, variance:
yS
tS
2
sS
=
1
n
yi
iS
STATISTICS 344, Lecture #7
Ratio Estimation
September 23, 2015
Scenario: You have been asked to estimate the number of
dairy cows in a regional district
Dairy farms in this district must be licensed, which gives you
access to a list of all them - a sampli
STATISTICS 344, Lecture #6
Some of the math under the hood, and a
wrap-up of Unit 1
September 21, 2015
Now what about some math.where does this tweaked
theory of simple random sampling come from?
Aim for avour on this (if taking 302 concurrently, the foll
STATISTICS 344, Lecture #8
Ratio Estimation, Continued
September 28, 2015
Recall ratio estimation idea: Y of interest, X auxiliary
Record both (xi , yi ) for sampled units (for i S, in our notation).
Target of estimation is yP , but happen to know xP (say
WENCHEN GUAN
37748143
STAT 344 Assignment 2
Chapter 3
3.2
a) Verbal: mean = i =508.33
Variance Sy2= = 6246.64
Math: mean X-bar= I = 635.33
Variance Sx2 = = 6163.68
Covariance Sxy =[ ]/(30-1)= = 2671.76
Correlation = = = 0.43
b)
Math
400 to 520
520 to 640
Lexi Zhang
Assignment HW03 due 10/27/2015 at 06:00pm PDT
STAT344-101 2015W1
Solution: (a) Under simple random sampling, the sample
mean of the scores, yS = 71.18, estimates the population mean.
If presented with a dataset, it can be dangerous to try to an
Lexi Zhang
Assignment HW04 due 11/20/2015 at 06:00pm PST
STAT344-101 2015W1
Greater Numbsville is comprised of three villages: Lower
Numbsville (LN, 500 residents), Old Numbsville (ON, 700 residents), and Upper Numbsville (UN, 900 residents). A crazed
sta
STAT 344 Lecture Notes- Week 1
Remark: This note is anticipated as a condensed version of the textbook
to be delivered as lectures. I will mark the relevant sections in the textbook
as much as possible.
1.0. Preambles
Example 1 I have a bottle containing
nite or not innite populations?
The line is not so clear in other examples.
The nite population may itself be dynamic. When the
unemployment information is needed, the population itself changes
with time.
This above notion works only if you can sample as
Week 2, Continue from Week2a slides.
Jiahua,Chen
Stratied,Simple,Random,Sample
2.3.1 Some notions the textbook omitted
The general task of a statistician is to make inference about the
stochastic system based on a data set generated from this system.
Othe
STAT 344 Lecture Notes- Week 4
Remark: This part of the note matches Chapter 4 and more.
4.1. Population proportion There is no doubt that proportion of units
with a specic characteristic in the population is an important measure. Proportion of female pro
STAT 344 Lecture Notes- Week 2
Remark: This part of the note matches Chapter 2 and more.
2.1. Finite population
Generally speaking, a target population in survey context are nite populations. If you keep taking sampling units from it without replacement,
Week 2, Continue from Week2b slides.
Jiahua,Chen
Week2c
2.3.2 Back to sampling plan SRSWOR
Suppose a nite population has N units, and a sampling plan
makes every subset of size n equally likely to become a sample.
In this case, the sampling plan is Simple
STAT 344 Lecture Notes- Week 3
Remark: This part of the note matches Chapter 3 and more.
3.1. Bias, variance and MSE of an estimator A statistical problem is
invariably to make inference on some aspects of a random system based on
the information collecte
STAT 344- Assignment 1
Marking scheme:
0: practically nothing sensible;
1: good eort; 2: sensible;
3: correct in principle; 4. nearly or is perfect.
Very simple problems worth 2 marks rather than 4. Extra length problems
will worth 8 marks.
Due date: Sept
STAT 344- Assignment 2
Marking scheme:
0: practically nothing sensible;
1: good eort; 2: sensible;
3: correct in principle; 4. nearly or is perfect.
Very simple problems worth 2 marks rather than 4. Extra length problems
will worth 8 marks.
Due date: Oct
Stat 344, Midterm Nov 14, 2014
Instructor: Jiahua Chen; Duration: 50mins; Aids: Calculators
1. [3] Provide one key dierence between concepts of stratum and cluster.
Ans: We do not sample strata but sample clusters.
2. [3] Consider the problem of ecient es
Lexi Zhang
Assignment HW02 due 10/09/2015 at 06:00pm PDT
STAT344-101 2015W1
The le lawyers.csv linked below contains the result of a
simple random sample of 35 out of 220 counties that comprise
the mythical land of Nod. Each row corresponds to a sampled c
STATISTICS 344, Lecture #11
Stratication I
(more or less, two loose ends from estimation in
domains to tie up rst)
October 5, 2015
Loose End #1: Recall the sample domain mean is really a
ratio estimator
uS
,
xS
where U and X have special denitions:
ySd
=
STATISTICS 344, Lecture #12
Stratication II
October 7, 2015
Recall example from last time - percentage of voters
supporting the federal Liberal party
Implement regional versions of the same study.
Region
Atlantic
Quebec
Ontario
West
Share of
population
6.
STAT 344: Nov. 13, 2015
First Thoughts about Missing Data
Say that a media report states that 60% of Canadians like maple syrup, with the results
considered accurate to within 3%, nineteen times of out twenty. Now, you dig deeper and
locate a more detaile
STATISTICS 344, Lecture #11
Estimation in Domains
September 30, 2016
Preamble
You are given the following data: Amongst 116 male voters
questioned, 34 indicated support for the NDP, while the other 82
indicated support for other parties. Thus you issue a
STATISTICS 344, Lecture #13
Stratification II
October 5, 2016
Recall example from last time - percentage of voters
supporting the federal Liberal party
Implement regional versions of the same study.
Region
Atlantic
Quebec
Ontario
West
Share of
population
STATISTICS 344, Lecture #12
Stratification I
October 3, 2016
Today we look at the reverse situation from last time
Estimation in a domain: Take a single sample from the whole
population. Use the portion of the sample that happens to fall in
the sub-popula
STATISTICS 344, Lecture #6
Sample Surveys and Change Over Time
September 19, 2016
Motivation
Surveys of political opinion, especially during an election campaign,
are amongst the most visible applications of sample-surveys.
They can often trigger media he