ECO S394D: Probability and Statistics

Instructor: James Scott (james.scott@mccombs.utexas.edu)

Office: CBA 6.478
Office hours: T Th 9-10 a.m.

Lectures: M T W Th, 1:30 3:15 p.m., Burdine 212

TA session: Friday 1:30 2:30 p.m., CLA 1.106

TA: Rong Chen (rong.chen@utmail.utexas.edu)

TA office hours: TBA

Background and objectives. This course is an intermediate level introduction to

probability theory and statistics. The main goal is to prepare students for econometrics.
Students taking the course should have a working knowledge of calculus and basic
descriptive statistics. The emphasis will be on conveying the basic principles of statistical
theory. This is fast-paced course with a condensed schedule, so it is essential to attend
lectures regularly and keep up with the material.

Course materials. Your main point of reference will be the scribe notes prepared by your
classmates (see below). As a reference on many probability topics we will also use:
Introduction to Probability Lecture Notes, by Dimitri P. Bertsekas and John N. Tsitsiklis.
This is posted as a PDF file on the course Canvas page.

I will also draw on All of Statistics: A Concise Course in Statistical Inference by Larry
Wasserman. You should not feel obligated to buy this. It is entirely optional. As the
material were covering is very standard, basically any intermediate probability/statistics
textbook can be used as a supplement.

Communication. I will post all assignments, announcements, etc., on Canvas. It is your

responsibility to check this site regularly.

Software. While this is a fairly theoretical course, the homework will sometimes include
some simulation or data-analysis exercises. These are all doable in Excel, but you can also
use R, Stata, Matlab anything youre comfortable with.

Assignments, exams and grading. There are five components to your grade: homeworks
(30%), midterm (15%), final exam (30%), probability in the real world (15%), and scribing
(10%).
There will be three problem sets, worth 10% each. These will be posted to Canvas (along
with their due dates).

There will be an in-class mid-term exam on Thursday, July 27th, worth 15%. The TA session
on the following day will be devoted to going over the mid-term.

The final exam will take place in class on Monday, August 14th, and is worth 30%.

I am also asking everyone to complete a probability in the real world writing assignment,
worth 15% and due on Thursday, August 3rd. This will involve both a written report and a
short (just a few minutes) in-class presentation that Thursday. I will give details on this
assignment in class, but the gist of it is simple: conduct a case study of some company,
organization, person, government, etc. using probability in pursuit of a larger economic,
research, or policy goal. Explain how theyre using probability to do something cool. The
idea is for you to connect the concepts we learn in class to some bigger picture in the world
of applications. I will give examples of this kind of thing in lectures.

Finally, the last 10% of your grade will come from submitting a high-quality scribe report
for a single lecture. I'll explain more about this in class, but the idea is simple: each day,
someone takes detailed notes and makes them available to their classmates. These notes are
supposed to be definitive, by which I mean: comprehensive, mistake-free, and easily
understood by any of your classmates who paid attention that day. You only have to do this
once, but you'll be repaid with a set of high-quality crowd-sourced lecture notes. I'll circulate
a sign-up sheet for scribing on the first day of class. You will e-mail your scribe notes to the
TA before the beginning of class the next day, who will post them in a folder on Canvas for
everyone to use.

Course outline
I. Probability Theory
Introduction: Random experiment. Set operations, Kolmogorovs rules of probability.
Interpretations of probability. Probability and coherence. Basic counting rules.
Joint, conditional, and marginal probability. Rule of total probability. Independence,
compounding, and confounding. Bayes rule.
Random variables and their distribution. Discrete random variables and their PMFs:
binomial, Poisson. Continuous random variables and the CDF/PDF: normal
distribution. Transformations of random variables.
 The bivariate normal distribution. Regression to the mean.
Properties of distributions: expectation (mean), variance, higher moments, conditional
expectation, conditional variance, covariance and correlation.

II. Statistical Theory: point estimation and the evaluation of estimators

Random sampling, i.i.d. random variables. Population parameters and sample statistics
(estimators). The concept of a sampling distribution.
General methods for constructing estimators: maximum likelihood, method of moments,
generalized method of moments, least squares. (Maybe a brief mention of Bayes
estimators.)
Evaluating estimators: finite sample properties. The exact sampling distribution of some
estimators. Unbiasedness. Relative efficiency. Mean squared error.
Evaluating estimators: asymptotic (large sample) properties. Convergence in
probability. The Law of Large Numbers. Convergence in distribution. The Central
Limit Theorem. Consistency. Deriving the asymptotic distribution of estimators.
Asymptotic normality, efficiency.

III. Statistical Theory: interval estimation and testing

Confidence intervals.
Hypothesis testing: basic concepts. Null and alternative hypotheses, confusion matrix,
Type 1 and 2 errors, power function. Constructing test statistics from estimators,
steps of implementing a test, p-values.
Time permitting: Intro to some more advanced concepts in hypothesis testing: e.g.,
likelihood ratio tests, NeymanPearson lemma, uniformly most powerful tests, large
sample tests (likelihood ratio and Wald tests)