Top Banner
The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007
32

The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

Dec 20, 2015

Download

Documents

matching slide

average treatment effect

instrument circle relative

education blue

random assignment

average untreated measures

weak instruments

selection effect

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

The Fundamental Problem of Causal Inference

Alexander Tabarrok

January 2007

Page 2: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 3: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 4: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 5: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 6: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

The Gold Standard• The gold standard is randomization. If units are randomly

assigned to treatment then the selection effect disappears.

• i.e. with random assignment the groups selected for treatment and the groups actually treated would have had the same outcomes on average if not treated.

• With random assignment the average treated minus the average untreated measures the average treatment effect on the treated (and in fact with random assignment this is also equal to the average treatment effect).

• In a randomized experiment we select N individuals from the population and randomly split them into two groups the treated with Nt members and the untreated with N-Nt.

Page 7: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

• In a regression context we can run the following regression:

• and Bt will measure the treatment effect. It's useful to run through this once in the simple case to prove that this is true. See handout.

Yi BT Ti i

Page 8: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

Matching

Page 9: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 10: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

Propensity Score0 1

E(Wages| _)Union

Non-Union

Wages of Union and Non-Union Members Against PropensityScore (Prob. of belonging to Union)

Effect of Observableson Wages for UnionMembers

Effect of Observableson Wages for NonUnion Members

TreatmentEffect

Page 11: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 12: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

The circle labeled "earnings" illustrates variation in the variable to be explained. Education and Ability are correlated explanatory variables and Ability is not observed. The blue area within the instrument circle represents variation in education that is uncorrelated with Ability and which can be used to consistently estimate the coefficient on education. Note that the only reason the instrument is correlated with Earnings is through education.

Page 13: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

Instruments in Action(Angrist and Krueger 1991)

Page 14: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

Instrumental variables with weak instruments and correlation with unobserved influences.

Bias in the IV estimator is determined by the covariance of the instrument with education (blue within instrument circle) relative to the covariance between the instrument and the unobserved factors (red within instrument circle). Thus IV with weak instruments can be more biased than OLS.

Page 15: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

2030

4050

Vio

len

t Cri

me

Rat

e

400

600

800

100

01

200

140

0P

riso

n P

op

ula

tion

(10

00s)

1980 1985 1990 1995 2000 2005Year...

Prison Population (1000s) Violent Crime Rate

Violent Crime Rate is violent crimes per 1000 pop over the age of 12Source: Bureau of Justice Statistics, http://www.ojp.usdoj.gov/bjs/

Page 16: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 17: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 18: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 19: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 20: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 21: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 22: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 23: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 24: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 25: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 26: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 27: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 28: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 29: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 30: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 31: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.
Page 32: The Fundamental Problem of Causal Inference Alexander Tabarrok January 2007.

From Did Securitization Lead to Lax Screening? Evidence From Subprime Loans (Benjamin J. Keys, Tanmoy Mukherjee, Amit Seru, Vikrant Vig)

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1093137

Related Documents
CAUSAL INFERENCE IN STATISTICS

CAUSAL INFERENCE IN STATISTICS

Category: Documents
Causal Inference

Causal Inference

Category: Documents
Causal Inference and Machine Learning - uni-potsdam.de · 2019-09-09 · Causal Inference Course 1 September 2019 Potsdam Causal Inference and Machine Learning Guido Imbens, imbens@stanford.edu

Causal Inference and Machine Learning - uni-potsdam.de ·.....

Category: Documents
Principles of Causal Inference

Principles of Causal Inference

Category: Documents
Generalizability in Causal Inference - Carlos Cinelli and Bareinboim...Cinelli, Bareinboim, SoCal 2019 - Generalizability in Causal Inference Causal inference: causal models Inputs:

Generalizability in Causal Inference - Carlos Cinelli and...

Category: Documents
Origo: causal inference by compression · 2018. 6. 29. · Origo: causal inference by compression 287 – a practical framework for causal inference based on MDL, – a causal inference

Origo: causal inference by compression · 2018. 6. 29. ·....

Category: Documents
Introduction to Causal Bayesian Inference Chris Holmes ... · Causal Bayesian Inference 3 Overview of Course – Introduction to causal inference and statistical association – The

Introduction to Causal Bayesian Inference Chris Holmes...

Category: Documents
Causal inference

Causal inference

Category: Documents
Causal Inference v16

Causal Inference v16

Category: Documents
Counterfactuals and Causal Inference – II

Counterfactuals and Causal Inference – II

Category: Documents
Causal inference in practice

Causal inference in practice

Category: Science
Causal inference in data science

Causal inference in data science

Category: Data & Analytics