Lecture 4 - Roadmap
Last Lecture, we talked about the linear regression model with
cross-sectional data and asked when the coecients in the linear
regression model have a causal interpretation.
Causal Eects and Selection Bias
Causality and Randomized Con

ARE 107 Problem Set 1
Due January 26, 2017 in Class
Prepared by Dalia Ghanem and Sophie Xu
Note: You are welcome to work in groups of a maximum of three students. No late assignment will be
accepted.
1. The CEO of a grocery food chain in California would

Lecture 4 with answers
ARE 107: Econometrics for Business Decisions
Sophie. Xu
1/19/2017
1 / 16
Announcement
I
Problem set 1 due Jan 26 in class.
Note: You are welcome to work in groups of a maximum of
three students. No late assignment will be accepted.

ARE 107: Econometrics for Business Decisions
Lecture 3
Sophie. Xu
UC Davis
Recap
New office hour: Tuesday 12-1pm, and Thursday 9-10am. Ok?
Any question so far?
= =
= ( = = ) + ( = = )
Average Causal Effect
=
Selection Bias
vHere is what would have h

Lecture 5 with answers
ARE 107: Econometrics for Business Decisions
Sophie. Xu
25 January, 2017
1 / 17
Announcement
I
Typo in the Problem Set 1. Corrected Problem set 1 is now
available on Smartsite.
Question 1, part (e)
If we estimate the model in part d

Lecture 5
ARE 107: Econometrics for Business Decisions
Sophie. Xu
24 January, 2017
1 / 16
Announcement
I
Typo in the Problem Set 1. Corrected Problem set 1 is now
available on Smartsite.
Question 1, part (e)
If we estimate the model in part d using OLS.
I

ARE 107 Lecture Notes
Lecture 1
Dalia Ghanem, UC Davis WELCOME TO ARE 107
How does it build on ARE 106?
ARE 106 is a course in linear regression with cross—sectional data, e.g. a
random sample of individuals, ﬁrms.
ARE 107 is different from ARE 106 in two

ARE 107 Lecture Notes
Lecture 3
Dalia Ghanem, UC Davis
1
Roadmap
So far.
I
Predictive vs. Causal Modeling
I
Randomized Control Trials
Causal Effect: E [Y1i Y0i |Di = 1]
The conditional expectation given Di = 1 means we are taking the average over
individu

ARE 107 Solutions to In-class Exercises
Dalia Ghanem, UC Davis
1
Lecture 1
Exercise 1.
Question 1: Y is quantity of orange juice sold, X is 1 if it is a rainy week
and 0 otherwise. We want to estimate the E [Y |X = 1] is the average
quantity sold given th

Lecture 2 - Roadmap
Causal vs. Predictive Modeling
Causal Inference: Randomized Control Trials (RCTs)
Reading: Chapter 1 in Mastering Metrics
LAST LECTURE: potential outcomes notation, denition of causal
eect, and introduction of selection bias
TODAY:
- F

Lecture 5 - Roadmap
So far.
Randomized Control Trials: Random Assignment Eliminates Selection
Bias
Linear Model with Observation Data: Exogeneity Eliminates Selection
Bias
This week.
Linear Instrumental Variables Model
Reading: Chapter 3: Mastering Met

UC Davis
Dalia Ghanem
ARE 107 Problem Set 1
Due April 19, 2016 in Class
Late Assignments Not Accepted!
Instructions. You are welcome to work in groups of a maximum of three students. Each group
member has to hand in their individual problem sets with the

Lecture 6 - Roadmap
This week.
Linear Instrumental Variables Model
Reading: Chapter 3: Mastering Metrics
- Lecture 5: Treatment Variable
- Today: Linear Model with Continuous Regressor
c
Dalia
Ghanem
16
RECAP: What have we done last lecture?
17
Linear In

ARE 107: Econometrics For Business Decisions
University of California, Davis
Spring 2016 Syllabus
Instructor:
TA:
Class Website:
Lectures:
Section:
Assignments:
Exams:
Grading:
Dalia Ghanem, OH: Tue and Thu 3:30-4:30pm in SS&H 3107.
Oscar Barriga Cabanill

UC Davis
Dalia Ghanem
ARE 107 Formula Sheet
Midterm 1
If Yi is an outcome variable and Di is a treatment variable,
E rYi |Di
If Yi
OLS
If Yi
1s E rYi|Di 0s E rY1i Y0i|Di 1s
Xi
ui ,
Cov pXi , ui q
V arpXi q
Xi
Wi
SR LR
E rY0i |Di
i ,
pXi, Wiq
Cov
V arpX

UC Davis
Dalia Ghanem
ARE 107 Problem Set 1
Answer Key
I. (a) The key issue in comparing the differences in average sales of those stores that opted
for the discount and those that did not is selection bias. In class, we decomposed this
difference in aver

ARE 107: Econometrics for Business Decisions
Lecture 2
Sophie. Xu
UC Davis
Roadmap
Syllabus
Causal vs. Predictive Modeling
Differentiate causal modeling from predictive modeling
From a decision-making perspective:
-Its appropriate to use_, if people i