I NTRODUCTION TO E CONOMIC S TATISTICS

(경제통계분석)
I NTRODUCTION

노승화
S EPTEMBER 2, 2024
Welcome to Intro to Economic Statistics class!

1 / 23
Instructor : Seunghwa Rho (노승화)
• [email protected]
• 경금대 610
• OH : Wednesday 5:20-6:00p(office) & or by appointment

2 / 23
W HY DO WE LEARN STATISTICS / ECONOMETRICS ?

• We often want to answer questions using data.

✓ Effectiveness of child safety seats?
✓ Effectiveness of a job training program?
✓ Impact of having an access to good health care system
• When answering questions, often what we want is the
causal relationship not correlation
The importance of holding other things fixed
The meaning of holding other things fixed and
random experiment

3 / 23
• Even after recovering causal relationship, we cannot say
something meaningful by just using the estimate itself.
✓ Suppose that having an access to good health care system
increased overall health score by 1.5. Would this imply that
having an access to good health care system indeed
improves health?
• We need to know where the estimate stands with our
assumption and find out if the estimate supports our
assumption. To be able to do this, we need to know the
distribution of the estimator with our assumption.
• There are various uncertainties (causal or not, asymptotics,
type I error) and this is why we learn econometrics. It is
not an easy job.
• This implies that it is a very useful subject!

4 / 23
W E OBSERVE SAMPLE NOT POPULATION

• If you are able to get the height for all the students at
Emory, then you can obtain the average. However, it is
likely that you would not have height information for all
the Emory students.
• It is more likely that you would have information on a
subset of Emory students.
• Here, all the Emory students are population, the true
world that you want to know. The subset of Emory
students that you have height information is sample.

5 / 23
To answer a research question, you identify the population of
interest from which you will collect your sample data.
• A population is the set of all subjects of interest.
• A sample is the subset of the population of interest on
which you collect data.
population

sample

6 / 23
• It is important that your sample is indeed sampled from
the population of interest.
• For example, when collecting income information,
high-income earner has a tendency not to respond to the
survey. In this case, due to the non-responses, your sample
may not be representative of population.

7 / 23
• In addition to your sample coming from the population of
interest, the observations would be independently.
• In summary, your observations are independent and
sampled from the same population of interest. This means
you have a random sample.

8 / 23
O THER SAMPLING METHODS

• Stratified sampling is a divide-and-conquer sampling

strategy.
• The population is divided into groups called strata. The
strata are chosen so that similar cases are grouped together.
• Next, second sampling method, usually simple random
sampling, is employed within each stratum.

9 / 23
• In a cluster sample, we break up the population into many
groups, called clusters. Then we sample fixed number of
clusters and include all observations from each of those
clusters in the sample.
• A multistage sample is like a cluster sample, but rather
than keeping all observations in each cluster, we collect a
random sample within each selected cluster.

10 / 23
O THER SAMPLING METHODS

top: random sample bottom: stratified sample

11 / 23
O THER SAMPLING METHODS

top: cluster sample bottom: multistage sample

12 / 23
• In addition, knowing some computer science algorithm
helps.
✓ New type of data appeared such as text data.
✓ Big data
✓ Other merits
• Some other database related knowledge helps such as SQL.
• Critical thinking, theoretical knowledge, and
computational skills are equally important!

13 / 23
 polviews >= 4.1
< 4.1

hrs1 >= 52 educ < 16

< 52 >= 16

marital < 2
>= 2

−0.4 −0.42 −0.42 −0.32 −0.28

7% 29% 20% 22% 21%

14 / 23
• This class is a baby step for your journey of learning how
to answer questions using data.
• I want all of you
✓ to understand statistical theory covered in class (of course).
However, I don’t want you to approach this technically
(theoretically) only. You should also understand intuitively
so that you can explain to someone who knows nothing
about statistics.
✓ Computational skills through statistical software are as
important as a theoretical and intuitive understanding of
statistical theory. Without computational skills, even if you
know the theory, you cannot analyze the data! This is the
case for whatever class you take related to data science.

15 / 23
S YLLABUS RELATED - 1. C LASS M ATERIAL

• Textbook
Openintro statistics 4th edition
https://www.openintro.org/book/os/
• Download and install R.
✓ MAC https://cran.r-project.org/bin/macosx/
✓ Windows https://cran.r-project.org/bin/windows/base/
• Download and install R studio
(It should be installed after you install R)
https://posit.co/download/rstudio-desktop/
• Datacamp : Link would be provided later through LMS

16 / 23
17 / 23
18 / 23
19 / 23
S YLLABUS RELATED - 2. G RADE

Grade is based on
• Homework (40%)- the lowest one would be dropped
• Midterm Exam (30%)
• Final Exam (30%)

20 / 23
S YLLABUS RELATED - 3. C OURSE C ONTENT AND
C OURSE O BJECTIVE

• Students are able to describe data through plots and

summary statistics.
• Students can answer questions using data through
hypothesis test.
• Students have a solid introduction to R.

21 / 23
S YLLABUS RELATED - 4. D ISABILITY
A CCOMMODATIONS

• If you are seeking academic adjustments, please discuss

with me or the support center for students with disabilities
ASAP. Support Center for Students with Disabilities
(Seoul) 02-2220-0776
• Students with visual impairments, physical disability,
developmental and intellectual disabilities, or other
matters can get assistance related to class registration
support, writing (note takers) support, extension of the test
time, enlarged textbook, etc

22 / 23
H OMEWORK

• Please install R, R studio

• Make sure to get an access to DataCamp using
HYU email address.

23 / 23