Simple Linear Regression

Size: px
Start display at page:

Download "Simple Linear Regression"

Transcription

1 Simple Linear Regression Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

2 Linear Regression and Causality Regression conditional expectation function of Y given X E(Y X) = f (X) = β X Question: When can we interpret coefficients as causal effects? Causal model Structural equation model Y i (t) = α + βt + ɛ i for t = 0, 1, where E(ɛ i ) = 0 1 No interference between units 2 α = E(Y i (0)) 3 β = Y i (1) Y i (0) for all i Constant additive unit causal effect Heterogenous treatment effect model: Y i (t) = α + β i t + ɛ i = α + βt + ɛ i (t) where E(ɛ i ) = 0 and ɛ i(t) = ɛ i + (β i β)t for t = 0, 1 ATE: β = E(β i ) = E(Y i (1) Y i (0)) E(ɛ i (t)) = 0 for t = 0, 1 and α = E(Y i (0)) Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

3 Identification Assumption Strict exogeneity assumption: E(ɛ i T) = E(ɛ i ) = 0 where T = (T 1, T 2,..., T n ) Under this assumption, least squares estimate ˆβ is unbiased for β Randomization of treatment: {Y i (1), Y i (0)} n T E(Y i (t) T) = E(Y i (t)) E(ɛ i (t) T) = E(ɛ i (t)) = 0 E(Y i (t)) = E(Y i T) = α + βt i Random sampling of units: {Y i (1), Y i (0)} {Y j (1), Y j (0)} for any i j {ɛ i (1), ɛ i (0)} {ɛ j (1), ɛ j (0)} for any i j Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

4 Unbiasedness of Least Squares Estimator Let (ˆα, ˆβ) be the least squares estimators When the treatment is binary, ˆα = 1 n (1 T i )Y i n 0 ˆβ = 1 n T i Y i 1 n 1 n 0 n (1 T i )Y i So, ˆα and ˆβ are unbiased for E(Y (0)) and E(Y i (1) Y i (0)) from the randomization inference perspective The same conclusion holds if T is categorical When T is continuous, the model assumes Y i (t) is linear in T Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

5 Homoskedasticity Homoskedastic error: V(ɛ T) = σ 2 I n 1 Random sampling of units implies ɛ i ɛ j 2 Equal variance of potential outcomes V(ɛ i (t)) = V(ɛ i (t) T) = V(Y i (t) T) = V(Y i (t)) = σ 2 t for t = 0, 1 Under the homoskedasticity assumption, model-based variance is, V( ˆβ T) = σ 2 n (T i T ) 2 standard model-based variance estimator is, V( ˆβ T ) = ˆσ 2 n (T i T ) 2 where ˆσ 2 = 1 n 2 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18 n ˆɛ 2 i

6 Violation of the Homoskedasticity Assumption Homoskedasticity assumption may be unrealistic: σ1 2 σ2 0 ( ) ( ) ˆσ 2 σ1 2 Bias = E n (T i T ) 2 Bias is zero when }{{} under complete randomization = (n 1 n 0 )(n 1) (σ1 2 n 1 n 0 (n 2) σ2 0 ) 1 homoskedasticity assumption holds: σ 2 1 = σ2 0 2 design is balanced: n 1 = n 0 Bias can be negative or positive + σ2 0 n 1 n 0 }{{} true variance Bias is typically small but does not go away asymptotically Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

7 Heteroskedasticity-Robust Variance Estimator Eicker-Huber-White (EHW) robust variance estimator: V EHW ((ˆα, ˆβ) X) }{{} sandwich = (X X) 1 (X diag(ˆɛ 2 i }{{}}{{ )X) (X X) 1 }}{{} = bread meat where X i = (1, T i ) and X = (X 1,..., X n ) When the treatment is binary, bread ( n ) 1 ( n ) ( n X i X i ˆɛ 2 i X ix i X i X i V EHW ( ˆβ T) = σ2 1 n 1 + σ2 0 n 0 where σ 2 t = 1 n t ( ) Bias = E( V( σ1 ˆβ 2 T)) + σ2 0 n 1 n 0 ) 1 n 1{T i = t}(y i Y t ) 2 = ( ) σ1 2 n1 2 + σ2 0 n0 2 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

8 Improved Robust Variance Estimators HC2 heteroskedasticity-robust variance estimator: ( ) } V HC2 ((ˆα, ˆβ) X) = (X X) {X 1 ˆɛ 2 i diag X (X X) 1 1 p i where p i = X i (X X) 1 X i is the leverage When the treatment is binary, V HC2 ( ˆβ T) = ˆσ2 1 n 1 + ˆσ2 0 n 0 Behrens-Fisher problem: even under normality, ( ˆβ β)/ VHC2 ( ˆβ T) does not follow the Student s t-distribution Degrees of freedom adjustments by Welch (1951) and Bell and McCaffery (2002) to improve approximation Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

9 Reducing Transphobia (Brookman and Kalla Science) Can perspective-taking imagining the world from another s perspective reduce transphobia? Treatment group (n 1 = 912): canvasser asking people to think about a time when they were judged unfairly and were guided to translate that experience to a transgender individual s experience Control group (n 0 = 913): canvasser asking people to recycle Outcome variable: feeling thermometer towards transgender people (0 100) 1 baseline 2 3 day 3 3 week 4 6 week 5 3 month Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

10 Durable Effects 1 3 day effect n 1 = 291, n 0 = 309, ˆσ 2 1 = , ˆσ2 0 = Neyman: est = 6.76, s.e. = 2.32, 95% CI = [2.23, 11.28] Regression: est = 6.76, s.e. = 2.30, 2.31 (HC), 2.32 (HC2) 2 3 month effect n 1 = 179, n 0 = 206, ˆσ 2 1 = , ˆσ2 0 = Neyman: est = 5.30, s.e. = 2.78, 95% CI = [ 0.15, 10.75] Regression: est = 5.30, s.e. = 2.78, 2.78 (HC), 2.78 (HC2) For simplicity, we ignored attrition, which can be problematic The treatment group has more attrition and a larger variance Relatively balanced sample homoskedasticity assumption matters little Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

11 Cluster Randomized Experiments Setup: Clusters of units: j = 1, 2,..., J Number of treated (control) clusters: J 1 (J 0 ) Units within cluster j: i = 1, 2,..., m j Total sample size: n = J j=1 m j Treatment assignment at cluster level: T j {0, 1} Outcome: Y ij = Y ij (T j ) Assumptions: 1 Random assignment: {Y ij (1), Y ij (0)} T j and Pr(T j = 1) = J 1 /J often easy to assign to each cluster 2 No interference across clusters useful when interference within a cluster is likely Causal quantities of interest: SATE 1 n m J j (Y ij (1) Y ij (0)) j=1 PATE E(Y ij (1) Y ij (0)) Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

12 Randomization Inference Assume equal cluster size for simplicity, i.e., m j = m for all j Unbiased estimator: ( ˆτ cluster = 1 J 1 T j J 1 m j=1 ) m Y ij 1 J 0 ( ) J 1 m (1 T j ) Y ij m Easy to show E(ˆτ cluster O) = SATE and thus E(ˆτ cluster ) = PATE j=1 Population variance (random sampling of clusters): V(ˆτ cluster O) V(ˆτ cluster ) = V(Y j(1)) J 1 + V(Y j(0)) J 0 where for t = 0, 1 Y j (t) = 1 m m Y ij (t) Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

13 Intracluster Correlation Coefficient (ICC) Comparison with the standard variance: V(Y j (t)) J t = = V(ˆτ) = σ2 1 J 1 m + σ2 0 J 0 m ( m ) 2 1 J t m 2 E Y ij (t) m µ t 1 m m J t m 2 E (Y ij (t) µ t ) 2 + (Y ij (t) µ t )(Y i j(t) µ t ) = σ2 t J t m {1 + (m 1)ρ t} typically i =1 i i σ 2 t J t m Suppose m = 101 and ρ t = 0.5 more than 50 times greater Random effect model: η j + ɛ ij where ρ = V(η j )/{V(η j ) + V(ɛ ij )} 0 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

14 Cluster Robust Variance Estimator Cluster robust variance estimator (Liang and Zeger Biometrika): 1 V J J J cluster ((ˆα, ˆβ) X) = X j X j X j ˆɛ j ˆɛ j X j X j X j j=1 j=1 where X j = (1 m j, T j ) and ˆɛ j = (ˆɛ 1j,..., ˆɛ mj j) Consistent as the number of clusters goes to infinity Bias adjustment a.k.a. CR2 (Bell and McCaffrey Surv. Methodol.); meat = j=1 J X j (I mj P Xj ) 1/2ˆɛ j ˆɛ j (I mj P Xj ) 1/2 X j j=1 where P Xj = X j (X X) 1 X j Degree of freedom adjustments are also available Implication: cluster standard errors by the unit of treatment assignment Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18 1

15 Effects of Clientalistic Campaign ( Wantchekon World Politics) Does clientalism affect voting behavior? Experiment involving 4 main candidates during the 2001 presidential election in Benin 8 non-competitive districts (out of 84) are selected Random assignment within each district: 1 one village: clientalistic campaign 2 one village: public policy campaign 3 the other villages: control group Many experimental villages were at least 25 miles apart: The risk of contagion was thereby minimized Post-election survey of 2344 registered voters to measure vote choice Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

16 Stratified Cluster Randomized Design Reg. Sample Sample Population District Candidate Exp. Villages Voters Size Mean Mean Kandi Kerekou clientelist (0) 0.81 public policy (.50) 0.60 control (.18) 0.75 Nikki Kerekou clientelist (.21) 0.90 public policy (.24) 0.85 control (.20) 0.82 Bembereke Lafia clientelist (.26) 0.94 public policy (.30) 0.93 control (.28) 0.74 Perere Lafia clientelist (.42) 0.81 public policy (.33) 0.25 control (.40) 0.58 Abomey Soglo clientelist (.13) 0.91 public policy (.13) 0.90 control (.15) 0.86 Ouidah Soglo clientelist (.25) 0.86 public policy (.26) 0.72 control (0.44) 0.64 Aplahoue Amoussou clientelist (.13) 0.87 public policy (.28) 0.77 control (.20) 0.72 Dogbo Amoussou clientelist (.48) 0.65 public policy (.50) 0.47 control (0.44) 0.84 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

17 Empirical Results Neyman public policy est s.e cluster assignment No No Yes Yes stratification No Yes No Yes clientalism est s.e cluster assignment No No Yes Yes stratification No Yes No Yes Regression public policy clientalism HC2 CR2 HC2 CR2 est s.e Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

18 Summary Simple linear regression Difference-in-means estimator Homoskedasticity implies the equal variance of potential outcomes Heteroskedasticity-robust variance estimator relaxes this assumption Cluster randomized experiments: 1 permits interference within cluster 2 less powerful when intracluster correlation is high 3 use cluster-robust standard variance estimator Suggested readings: 1 ANGRIST AND PISCHKE, Chapter 2 2 IMBENS AND RUBIN, Chapter 7 Kosuke Imai (Harvard) Simple Linear Regression Stat186/Gov2002 Fall / 18

Instrumental Variables

Instrumental Variables Instrumental Variables Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Noncompliance in Experiments Stat186/Gov2002 Fall 2018 1 / 18 Instrumental Variables

More information

Inference for Average Treatment Effects

Inference for Average Treatment Effects Inference for Average Treatment Effects Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Average Treatment Effects Stat186/Gov2002 Fall 2018 1 / 15 Social

More information

Stratified Randomized Experiments

Stratified Randomized Experiments Stratified Randomized Experiments Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Stratified Randomized Experiments Stat186/Gov2002 Fall 2018 1 / 13 Blocking

More information

Regression Discontinuity Designs

Regression Discontinuity Designs Regression Discontinuity Designs Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Regression Discontinuity Design Stat186/Gov2002 Fall 2018 1 / 1 Observational

More information

The Essential Role of Pair Matching in. Cluster-Randomized Experiments. with Application to the Mexican Universal Health Insurance Evaluation

The Essential Role of Pair Matching in. Cluster-Randomized Experiments. with Application to the Mexican Universal Health Insurance Evaluation The Essential Role of Pair Matching in Cluster-Randomized Experiments, with Application to the Mexican Universal Health Insurance Evaluation Kosuke Imai Princeton University Gary King Clayton Nall Harvard

More information

Noncompliance in Randomized Experiments

Noncompliance in Randomized Experiments Noncompliance in Randomized Experiments Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Noncompliance in Experiments Stat186/Gov2002 Fall 2018 1 / 15 Encouragement

More information

Weighting Methods. Harvard University STAT186/GOV2002 CAUSAL INFERENCE. Fall Kosuke Imai

Weighting Methods. Harvard University STAT186/GOV2002 CAUSAL INFERENCE. Fall Kosuke Imai Weighting Methods Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Weighting Methods Stat186/Gov2002 Fall 2018 1 / 13 Motivation Matching methods for improving

More information

Clustering as a Design Problem

Clustering as a Design Problem Clustering as a Design Problem Alberto Abadie, Susan Athey, Guido Imbens, & Jeffrey Wooldridge Harvard-MIT Econometrics Seminar Cambridge, February 4, 2016 Adjusting standard errors for clustering is common

More information

Randomized Experiments

Randomized Experiments Randomized Experiments Teppei Yamamoto Keio University Introduction to Causal Inference Spring 2016 1 Introduction 2 Identification 3 Basic Inference 4 Covariate Adjustment 5 Threats to Validity 6 Advanced

More information

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data?

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data? When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data? Kosuke Imai Princeton University Asian Political Methodology Conference University of Sydney Joint

More information

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Panel Data?

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Panel Data? When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Panel Data? Kosuke Imai Department of Politics Center for Statistics and Machine Learning Princeton University Joint

More information

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data?

When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data? When Should We Use Linear Fixed Effects Regression Models for Causal Inference with Longitudinal Data? Kosuke Imai Department of Politics Center for Statistics and Machine Learning Princeton University

More information

Statistical Analysis of Randomized Experiments with Nonignorable Missing Binary Outcomes

Statistical Analysis of Randomized Experiments with Nonignorable Missing Binary Outcomes Statistical Analysis of Randomized Experiments with Nonignorable Missing Binary Outcomes Kosuke Imai Department of Politics Princeton University July 31 2007 Kosuke Imai (Princeton University) Nonignorable

More information

Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies

Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies Kosuke Imai Department of Politics Princeton University November 13, 2013 So far, we have essentially assumed

More information

On the Use of Linear Fixed Effects Regression Models for Causal Inference

On the Use of Linear Fixed Effects Regression Models for Causal Inference On the Use of Linear Fixed Effects Regression Models for ausal Inference Kosuke Imai Department of Politics Princeton University Joint work with In Song Kim Atlantic ausal Inference onference Johns Hopkins

More information

Robust Standard Errors in Small Samples: Some Practical Advice

Robust Standard Errors in Small Samples: Some Practical Advice Robust Standard Errors in Small Samples: Some Practical Advice Guido W. Imbens Michal Kolesár First Draft: October 2012 This Draft: December 2014 Abstract In this paper we discuss the properties of confidence

More information

AN EVALUATION OF PARAMETRIC AND NONPARAMETRIC VARIANCE ESTIMATORS IN COMPLETELY RANDOMIZED EXPERIMENTS. Stanley A. Lubanski. and. Peter M.

AN EVALUATION OF PARAMETRIC AND NONPARAMETRIC VARIANCE ESTIMATORS IN COMPLETELY RANDOMIZED EXPERIMENTS. Stanley A. Lubanski. and. Peter M. AN EVALUATION OF PARAMETRIC AND NONPARAMETRIC VARIANCE ESTIMATORS IN COMPLETELY RANDOMIZED EXPERIMENTS by Stanley A. Lubanski and Peter M. Steiner UNIVERSITY OF WISCONSIN-MADISON 018 Background To make

More information

Quantitative Economics for the Evaluation of the European Policy

Quantitative Economics for the Evaluation of the European Policy Quantitative Economics for the Evaluation of the European Policy Dipartimento di Economia e Management Irene Brunetti Davide Fiaschi Angela Parenti 1 25th of September, 2017 1 ireneb@ec.unipi.it, davide.fiaschi@unipi.it,

More information

NBER WORKING PAPER SERIES ROBUST STANDARD ERRORS IN SMALL SAMPLES: SOME PRACTICAL ADVICE. Guido W. Imbens Michal Kolesar

NBER WORKING PAPER SERIES ROBUST STANDARD ERRORS IN SMALL SAMPLES: SOME PRACTICAL ADVICE. Guido W. Imbens Michal Kolesar NBER WORKING PAPER SERIES ROBUST STANDARD ERRORS IN SMALL SAMPLES: SOME PRACTICAL ADVICE Guido W. Imbens Michal Kolesar Working Paper 18478 http://www.nber.org/papers/w18478 NATIONAL BUREAU OF ECONOMIC

More information

Identification and Inference in Causal Mediation Analysis

Identification and Inference in Causal Mediation Analysis Identification and Inference in Causal Mediation Analysis Kosuke Imai Luke Keele Teppei Yamamoto Princeton University Ohio State University November 12, 2008 Kosuke Imai (Princeton) Causal Mediation Analysis

More information

Introduction to Statistical Inference

Introduction to Statistical Inference Introduction to Statistical Inference Kosuke Imai Princeton University January 31, 2010 Kosuke Imai (Princeton) Introduction to Statistical Inference January 31, 2010 1 / 21 What is Statistics? Statistics

More information

Statistical Analysis of Causal Mechanisms

Statistical Analysis of Causal Mechanisms Statistical Analysis of Causal Mechanisms Kosuke Imai Princeton University November 17, 2008 Joint work with Luke Keele (Ohio State) and Teppei Yamamoto (Princeton) Kosuke Imai (Princeton) Causal Mechanisms

More information

ESTIMATION OF TREATMENT EFFECTS VIA MATCHING

ESTIMATION OF TREATMENT EFFECTS VIA MATCHING ESTIMATION OF TREATMENT EFFECTS VIA MATCHING AAEC 56 INSTRUCTOR: KLAUS MOELTNER Textbooks: R scripts: Wooldridge (00), Ch.; Greene (0), Ch.9; Angrist and Pischke (00), Ch. 3 mod5s3 General Approach The

More information

The Simple Linear Regression Model

The Simple Linear Regression Model The Simple Linear Regression Model Lesson 3 Ryan Safner 1 1 Department of Economics Hood College ECON 480 - Econometrics Fall 2017 Ryan Safner (Hood College) ECON 480 - Lesson 3 Fall 2017 1 / 77 Bivariate

More information

Applied Statistics Lecture Notes

Applied Statistics Lecture Notes Applied Statistics Lecture Notes Kosuke Imai Department of Politics Princeton University February 2, 2008 Making statistical inferences means to learn about what you do not observe, which is called parameters,

More information

Use of Matching Methods for Causal Inference in Experimental and Observational Studies. This Talk Draws on the Following Papers:

Use of Matching Methods for Causal Inference in Experimental and Observational Studies. This Talk Draws on the Following Papers: Use of Matching Methods for Causal Inference in Experimental and Observational Studies Kosuke Imai Department of Politics Princeton University April 27, 2007 Kosuke Imai (Princeton University) Matching

More information

The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors

The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors by Bruce E. Hansen Department of Economics University of Wisconsin October 2018 Bruce Hansen (University of Wisconsin) Exact

More information

Statistical Analysis of Causal Mechanisms

Statistical Analysis of Causal Mechanisms Statistical Analysis of Causal Mechanisms Kosuke Imai Luke Keele Dustin Tingley Teppei Yamamoto Princeton University Ohio State University July 25, 2009 Summer Political Methodology Conference Imai, Keele,

More information

Selection on Observables: Propensity Score Matching.

Selection on Observables: Propensity Score Matching. Selection on Observables: Propensity Score Matching. Department of Economics and Management Irene Brunetti ireneb@ec.unipi.it 24/10/2017 I. Brunetti Labour Economics in an European Perspective 24/10/2017

More information

Statistical Analysis of Causal Mechanisms

Statistical Analysis of Causal Mechanisms Statistical Analysis of Causal Mechanisms Kosuke Imai Princeton University April 13, 2009 Kosuke Imai (Princeton) Causal Mechanisms April 13, 2009 1 / 26 Papers and Software Collaborators: Luke Keele,

More information

Problems. Suppose both models are fitted to the same data. Show that SS Res, A SS Res, B

Problems. Suppose both models are fitted to the same data. Show that SS Res, A SS Res, B Simple Linear Regression 35 Problems 1 Consider a set of data (x i, y i ), i =1, 2,,n, and the following two regression models: y i = β 0 + β 1 x i + ε, (i =1, 2,,n), Model A y i = γ 0 + γ 1 x i + γ 2

More information

Causal Inference Lecture Notes: Selection Bias in Observational Studies

Causal Inference Lecture Notes: Selection Bias in Observational Studies Causal Inference Lecture Notes: Selection Bias in Observational Studies Kosuke Imai Department of Politics Princeton University April 7, 2008 So far, we have studied how to analyze randomized experiments.

More information

Use of Matching Methods for Causal Inference in Experimental and Observational Studies. This Talk Draws on the Following Papers:

Use of Matching Methods for Causal Inference in Experimental and Observational Studies. This Talk Draws on the Following Papers: Use of Matching Methods for Causal Inference in Experimental and Observational Studies Kosuke Imai Department of Politics Princeton University April 13, 2009 Kosuke Imai (Princeton University) Matching

More information

EMERGING MARKETS - Lecture 2: Methodology refresher

EMERGING MARKETS - Lecture 2: Methodology refresher EMERGING MARKETS - Lecture 2: Methodology refresher Maria Perrotta April 4, 2013 SITE http://www.hhs.se/site/pages/default.aspx My contact: maria.perrotta@hhs.se Aim of this class There are many different

More information

Combining multiple observational data sources to estimate causal eects

Combining multiple observational data sources to estimate causal eects Department of Statistics, North Carolina State University Combining multiple observational data sources to estimate causal eects Shu Yang* syang24@ncsuedu Joint work with Peng Ding UC Berkeley May 23,

More information

A Practitioner s Guide to Cluster-Robust Inference

A Practitioner s Guide to Cluster-Robust Inference A Practitioner s Guide to Cluster-Robust Inference A. C. Cameron and D. L. Miller presented by Federico Curci March 4, 2015 Cameron Miller Cluster Clinic II March 4, 2015 1 / 20 In the previous episode

More information

Final Exam. Economics 835: Econometrics. Fall 2010

Final Exam. Economics 835: Econometrics. Fall 2010 Final Exam Economics 835: Econometrics Fall 2010 Please answer the question I ask - no more and no less - and remember that the correct answer is often short and simple. 1 Some short questions a) For each

More information

Covariate Balancing Propensity Score for General Treatment Regimes

Covariate Balancing Propensity Score for General Treatment Regimes Covariate Balancing Propensity Score for General Treatment Regimes Kosuke Imai Princeton University October 14, 2014 Talk at the Department of Psychiatry, Columbia University Joint work with Christian

More information

Statistical Models for Causal Analysis

Statistical Models for Causal Analysis Statistical Models for Causal Analysis Teppei Yamamoto Keio University Introduction to Causal Inference Spring 2016 Three Modes of Statistical Inference 1. Descriptive Inference: summarizing and exploring

More information

Econometric Analysis of Cross Section and Panel Data

Econometric Analysis of Cross Section and Panel Data Econometric Analysis of Cross Section and Panel Data Jeffrey M. Wooldridge / The MIT Press Cambridge, Massachusetts London, England Contents Preface Acknowledgments xvii xxiii I INTRODUCTION AND BACKGROUND

More information

Econometrics. Week 4. Fall Institute of Economic Studies Faculty of Social Sciences Charles University in Prague

Econometrics. Week 4. Fall Institute of Economic Studies Faculty of Social Sciences Charles University in Prague Econometrics Week 4 Institute of Economic Studies Faculty of Social Sciences Charles University in Prague Fall 2012 1 / 23 Recommended Reading For the today Serial correlation and heteroskedasticity in

More information

Small sample methods for cluster-robust variance estimation and hypothesis testing in fixed effects models

Small sample methods for cluster-robust variance estimation and hypothesis testing in fixed effects models Small sample methods for cluster-robust variance estimation and hypothesis testing in fixed effects models arxiv:1601.01981v1 [stat.me] 8 Jan 2016 James E. Pustejovsky Department of Educational Psychology

More information

Recitation 5. Inference and Power Calculations. Yiqing Xu. March 7, 2014 MIT

Recitation 5. Inference and Power Calculations. Yiqing Xu. March 7, 2014 MIT 17.802 Recitation 5 Inference and Power Calculations Yiqing Xu MIT March 7, 2014 1 Inference of Frequentists 2 Power Calculations Inference (mostly MHE Ch8) Inference in Asymptopia (and with Weak Null)

More information

Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Causal Interaction in Factorial Experiments: Application to Conjoint Analysis Causal Interaction in Factorial Experiments: Application to Conjoint Analysis Naoki Egami Kosuke Imai Princeton University Talk at the University of Macau June 16, 2017 Egami and Imai (Princeton) Causal

More information

What s New in Econometrics. Lecture 1

What s New in Econometrics. Lecture 1 What s New in Econometrics Lecture 1 Estimation of Average Treatment Effects Under Unconfoundedness Guido Imbens NBER Summer Institute, 2007 Outline 1. Introduction 2. Potential Outcomes 3. Estimands and

More information

Causal Inference with General Treatment Regimes: Generalizing the Propensity Score

Causal Inference with General Treatment Regimes: Generalizing the Propensity Score Causal Inference with General Treatment Regimes: Generalizing the Propensity Score David van Dyk Department of Statistics, University of California, Irvine vandyk@stat.harvard.edu Joint work with Kosuke

More information

Causal Interaction in Factorial Experiments: Application to Conjoint Analysis

Causal Interaction in Factorial Experiments: Application to Conjoint Analysis Causal Interaction in Factorial Experiments: Application to Conjoint Analysis Naoki Egami Kosuke Imai Princeton University Talk at the Institute for Mathematics and Its Applications University of Minnesota

More information

Causal Mechanisms Short Course Part II:

Causal Mechanisms Short Course Part II: Causal Mechanisms Short Course Part II: Analyzing Mechanisms with Experimental and Observational Data Teppei Yamamoto Massachusetts Institute of Technology March 24, 2012 Frontiers in the Analysis of Causal

More information

Robustness to Parametric Assumptions in Missing Data Models

Robustness to Parametric Assumptions in Missing Data Models Robustness to Parametric Assumptions in Missing Data Models Bryan Graham NYU Keisuke Hirano University of Arizona April 2011 Motivation Motivation We consider the classic missing data problem. In practice

More information

Robust Estimation of Inverse Probability Weights for Marginal Structural Models

Robust Estimation of Inverse Probability Weights for Marginal Structural Models Robust Estimation of Inverse Probability Weights for Marginal Structural Models Kosuke IMAI and Marc RATKOVIC Marginal structural models (MSMs) are becoming increasingly popular as a tool for causal inference

More information

Notes on causal effects

Notes on causal effects Notes on causal effects Johan A. Elkink March 4, 2013 1 Decomposing bias terms Deriving Eq. 2.12 in Morgan and Winship (2007: 46): { Y1i if T Potential outcome = i = 1 Y 0i if T i = 0 Using shortcut E

More information

Implementing Matching Estimators for. Average Treatment Effects in STATA. Guido W. Imbens - Harvard University Stata User Group Meeting, Boston

Implementing Matching Estimators for. Average Treatment Effects in STATA. Guido W. Imbens - Harvard University Stata User Group Meeting, Boston Implementing Matching Estimators for Average Treatment Effects in STATA Guido W. Imbens - Harvard University Stata User Group Meeting, Boston July 26th, 2006 General Motivation Estimation of average effect

More information

Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies

Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies Kosuke Imai Princeton University Joint work with Keele (Ohio State), Tingley (Harvard), Yamamoto (Princeton)

More information

Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies

Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Unpacking the Black Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Kosuke Imai Department of Politics Center for Statistics and Machine Learning Princeton

More information

Statistical Analysis of Causal Mechanisms for Randomized Experiments

Statistical Analysis of Causal Mechanisms for Randomized Experiments Statistical Analysis of Causal Mechanisms for Randomized Experiments Kosuke Imai Department of Politics Princeton University November 22, 2008 Graduate Student Conference on Experiments in Interactive

More information

Identification, Inference, and Sensitivity Analysis for Causal Mediation Effects

Identification, Inference, and Sensitivity Analysis for Causal Mediation Effects Identification, Inference, and Sensitivity Analysis for Causal Mediation Effects Kosuke Imai Luke Keele Teppei Yamamoto First Draft: November 4, 2008 This Draft: January 15, 2009 Abstract Causal mediation

More information

The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors

The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors The Exact Distribution of the t-ratio with Robust and Clustered Standard Errors by Bruce E. Hansen Department of Economics University of Wisconsin June 2017 Bruce Hansen (University of Wisconsin) Exact

More information

Estimation of the Conditional Variance in Paired Experiments

Estimation of the Conditional Variance in Paired Experiments Estimation of the Conditional Variance in Paired Experiments Alberto Abadie & Guido W. Imbens Harvard University and BER June 008 Abstract In paired randomized experiments units are grouped in pairs, often

More information

Discussion: Can We Get More Out of Experiments?

Discussion: Can We Get More Out of Experiments? Discussion: Can We Get More Out of Experiments? Kosuke Imai Princeton University September 4, 2010 Kosuke Imai (Princeton) Discussion APSA 2010 (Washington D.C.) 1 / 9 Keele, McConnaughy, and White Question:

More information

Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies

Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Kosuke Imai Princeton University January 23, 2012 Joint work with L. Keele (Penn State)

More information

Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies

Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Unpacking the Black-Box of Causality: Learning about Causal Mechanisms from Experimental and Observational Studies Kosuke Imai Princeton University February 23, 2012 Joint work with L. Keele (Penn State)

More information

Data Integration for Big Data Analysis for finite population inference

Data Integration for Big Data Analysis for finite population inference for Big Data Analysis for finite population inference Jae-kwang Kim ISU January 23, 2018 1 / 36 What is big data? 2 / 36 Data do not speak for themselves Knowledge Reproducibility Information Intepretation

More information

Difference-in-Differences Methods

Difference-in-Differences Methods Difference-in-Differences Methods Teppei Yamamoto Keio University Introduction to Causal Inference Spring 2016 1 Introduction: A Motivating Example 2 Identification 3 Estimation and Inference 4 Diagnostics

More information

Causal Inference with Interference and Noncompliance in Two-Stage Randomized Experiments

Causal Inference with Interference and Noncompliance in Two-Stage Randomized Experiments Causal Inference with Interference and Noncompliance in Two-Stage Randomized Experiments Kosuke Imai Zhichao Jiang Anup Malani First Draft: April 9, 08 This Draft: June 5, 08 Abstract In many social science

More information

Potential Outcomes Model (POM)

Potential Outcomes Model (POM) Potential Outcomes Model (POM) Relationship Between Counterfactual States Causality Empirical Strategies in Labor Economics, Angrist Krueger (1999): The most challenging empirical questions in economics

More information

Identification and Sensitivity Analysis for Multiple Causal Mechanisms: Revisiting Evidence from Framing Experiments

Identification and Sensitivity Analysis for Multiple Causal Mechanisms: Revisiting Evidence from Framing Experiments Identification and Sensitivity Analysis for Multiple Causal Mechanisms: Revisiting Evidence from Framing Experiments Kosuke Imai Teppei Yamamoto First Draft: May 17, 2011 This Draft: January 10, 2012 Abstract

More information

ECON Program Evaluation, Binary Dependent Variable, Misc.

ECON Program Evaluation, Binary Dependent Variable, Misc. ECON 351 - Program Evaluation, Binary Dependent Variable, Misc. Maggie Jones () 1 / 17 Readings Chapter 13: Section 13.2 on difference in differences Chapter 7: Section on binary dependent variables Chapter

More information

1 Motivation for Instrumental Variable (IV) Regression

1 Motivation for Instrumental Variable (IV) Regression ECON 370: IV & 2SLS 1 Instrumental Variables Estimation and Two Stage Least Squares Econometric Methods, ECON 370 Let s get back to the thiking in terms of cross sectional (or pooled cross sectional) data

More information

Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies

Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies Unpacking the Black-Box: Learning about Causal Mechanisms from Experimental and Observational Studies Kosuke Imai Princeton University September 14, 2010 Institute of Statistical Mathematics Project Reference

More information

Simple Linear Regression

Simple Linear Regression Simple Linear Regression ST 430/514 Recall: A regression model describes how a dependent variable (or response) Y is affected, on average, by one or more independent variables (or factors, or covariates)

More information

Recent Advances in the Field of Trade Theory and Policy Analysis Using Micro-Level Data

Recent Advances in the Field of Trade Theory and Policy Analysis Using Micro-Level Data Recent Advances in the Field of Trade Theory and Policy Analysis Using Micro-Level Data July 2012 Bangkok, Thailand Cosimo Beverelli (World Trade Organization) 1 Content a) Classical regression model b)

More information

The propensity score with continuous treatments

The propensity score with continuous treatments 7 The propensity score with continuous treatments Keisuke Hirano and Guido W. Imbens 1 7.1 Introduction Much of the work on propensity score analysis has focused on the case in which the treatment is binary.

More information

Econometrics II. Nonstandard Standard Error Issues: A Guide for the. Practitioner

Econometrics II. Nonstandard Standard Error Issues: A Guide for the. Practitioner Econometrics II Nonstandard Standard Error Issues: A Guide for the Practitioner Måns Söderbom 10 May 2011 Department of Economics, University of Gothenburg. Email: mans.soderbom@economics.gu.se. Web: www.economics.gu.se/soderbom,

More information

The Multilevel Logit Model for Binary Dependent Variables Marco R. Steenbergen

The Multilevel Logit Model for Binary Dependent Variables Marco R. Steenbergen The Multilevel Logit Model for Binary Dependent Variables Marco R. Steenbergen January 23-24, 2012 Page 1 Part I The Single Level Logit Model: A Review Motivating Example Imagine we are interested in voting

More information

Heteroskedasticity ECONOMETRICS (ECON 360) BEN VAN KAMMEN, PHD

Heteroskedasticity ECONOMETRICS (ECON 360) BEN VAN KAMMEN, PHD Heteroskedasticity ECONOMETRICS (ECON 360) BEN VAN KAMMEN, PHD Introduction For pedagogical reasons, OLS is presented initially under strong simplifying assumptions. One of these is homoskedastic errors,

More information

Econometrics with Observational Data. Introduction and Identification Todd Wagner February 1, 2017

Econometrics with Observational Data. Introduction and Identification Todd Wagner February 1, 2017 Econometrics with Observational Data Introduction and Identification Todd Wagner February 1, 2017 Goals for Course To enable researchers to conduct careful quantitative analyses with existing VA (and non-va)

More information

Small-sample cluster-robust variance estimators for two-stage least squares models

Small-sample cluster-robust variance estimators for two-stage least squares models Small-sample cluster-robust variance estimators for two-stage least squares models ames E. Pustejovsky The University of Texas at Austin Context In randomized field trials of educational interventions,

More information

High Dimensional Propensity Score Estimation via Covariate Balancing

High Dimensional Propensity Score Estimation via Covariate Balancing High Dimensional Propensity Score Estimation via Covariate Balancing Kosuke Imai Princeton University Talk at Columbia University May 13, 2017 Joint work with Yang Ning and Sida Peng Kosuke Imai (Princeton)

More information

Advanced Quantitative Research Methodology, Lecture Notes: Research Designs for Causal Inference 1

Advanced Quantitative Research Methodology, Lecture Notes: Research Designs for Causal Inference 1 Advanced Quantitative Research Methodology, Lecture Notes: Research Designs for Causal Inference 1 Gary King GaryKing.org April 13, 2014 1 c Copyright 2014 Gary King, All Rights Reserved. Gary King ()

More information

Comprehensive Examination Quantitative Methods Spring, 2018

Comprehensive Examination Quantitative Methods Spring, 2018 Comprehensive Examination Quantitative Methods Spring, 2018 Instruction: This exam consists of three parts. You are required to answer all the questions in all the parts. 1 Grading policy: 1. Each part

More information

multilevel modeling: concepts, applications and interpretations

multilevel modeling: concepts, applications and interpretations multilevel modeling: concepts, applications and interpretations lynne c. messer 27 october 2010 warning social and reproductive / perinatal epidemiologist concepts why context matters multilevel models

More information

GLS and FGLS. Econ 671. Purdue University. Justin L. Tobias (Purdue) GLS and FGLS 1 / 22

GLS and FGLS. Econ 671. Purdue University. Justin L. Tobias (Purdue) GLS and FGLS 1 / 22 GLS and FGLS Econ 671 Purdue University Justin L. Tobias (Purdue) GLS and FGLS 1 / 22 In this lecture we continue to discuss properties associated with the GLS estimator. In addition we discuss the practical

More information

Least Squares Estimation-Finite-Sample Properties

Least Squares Estimation-Finite-Sample Properties Least Squares Estimation-Finite-Sample Properties Ping Yu School of Economics and Finance The University of Hong Kong Ping Yu (HKU) Finite-Sample 1 / 29 Terminology and Assumptions 1 Terminology and Assumptions

More information

Impact Evaluation Technical Workshop:

Impact Evaluation Technical Workshop: Impact Evaluation Technical Workshop: Asian Development Bank Sept 1 3, 2014 Manila, Philippines Session 19(b) Quantile Treatment Effects I. Quantile Treatment Effects Most of the evaluation literature

More information

Bootstrap-Based Improvements for Inference with Clustered Errors

Bootstrap-Based Improvements for Inference with Clustered Errors Bootstrap-Based Improvements for Inference with Clustered Errors Colin Cameron, Jonah Gelbach, Doug Miller U.C. - Davis, U. Maryland, U.C. - Davis May, 2008 May, 2008 1 / 41 1. Introduction OLS regression

More information

Why Do Statisticians Treat Predictors as Fixed? A Conspiracy Theory

Why Do Statisticians Treat Predictors as Fixed? A Conspiracy Theory Why Do Statisticians Treat Predictors as Fixed? A Conspiracy Theory Andreas Buja joint with the PoSI Group: Richard Berk, Lawrence Brown, Linda Zhao, Kai Zhang Ed George, Mikhail Traskin, Emil Pitkin,

More information

The Causal Inference Problem and the Rubin Causal Model

The Causal Inference Problem and the Rubin Causal Model The Causal Inference Problem and the Rubin Causal Model Lecture 2 Rebecca B. Morton NYU Exp Class Lectures R B Morton (NYU) EPS Lecture 2 Exp Class Lectures 1 / 23 Variables in Modeling the E ects of a

More information

Causal Directed Acyclic Graphs

Causal Directed Acyclic Graphs Causal Directed Acyclic Graphs Kosuke Imai Harvard University STAT186/GOV2002 CAUSAL INFERENCE Fall 2018 Kosuke Imai (Harvard) Causal DAGs Stat186/Gov2002 Fall 2018 1 / 15 Elements of DAGs (Pearl. 2000.

More information

Regression #3: Properties of OLS Estimator

Regression #3: Properties of OLS Estimator Regression #3: Properties of OLS Estimator Econ 671 Purdue University Justin L. Tobias (Purdue) Regression #3 1 / 20 Introduction In this lecture, we establish some desirable properties associated with

More information

Casuality and Programme Evaluation

Casuality and Programme Evaluation Casuality and Programme Evaluation Lecture V: Difference-in-Differences II Dr Martin Karlsson University of Duisburg-Essen Summer Semester 2017 M Karlsson (University of Duisburg-Essen) Casuality and Programme

More information

Panel Data Models. Chapter 5. Financial Econometrics. Michael Hauser WS17/18 1 / 63

Panel Data Models. Chapter 5. Financial Econometrics. Michael Hauser WS17/18 1 / 63 1 / 63 Panel Data Models Chapter 5 Financial Econometrics Michael Hauser WS17/18 2 / 63 Content Data structures: Times series, cross sectional, panel data, pooled data Static linear panel data models:

More information

Stat 579: Generalized Linear Models and Extensions

Stat 579: Generalized Linear Models and Extensions Stat 579: Generalized Linear Models and Extensions Linear Mixed Models for Longitudinal Data Yan Lu April, 2018, week 15 1 / 38 Data structure t1 t2 tn i 1st subject y 11 y 12 y 1n1 Experimental 2nd subject

More information

When Should We Use Fixed Effects Regression Models for Causal Inference with Longitudinal Data?

When Should We Use Fixed Effects Regression Models for Causal Inference with Longitudinal Data? When Should We Use Fixed Effects Regression Models for Causal Inference with Longitudinal Data? Kosuke Imai In Song Kim First Draft: July 1, 2014 This Draft: December 19, 2017 Abstract Many researchers

More information

Non-Spherical Errors

Non-Spherical Errors Non-Spherical Errors Krishna Pendakur February 15, 2016 1 Efficient OLS 1. Consider the model Y = Xβ + ε E [X ε = 0 K E [εε = Ω = σ 2 I N. 2. Consider the estimated OLS parameter vector ˆβ OLS = (X X)

More information

Matching and Weighting Methods for Causal Inference

Matching and Weighting Methods for Causal Inference Matching and Weighting Methods for ausal Inference Kosuke Imai Princeton University Methods Workshop, Duke University Kosuke Imai (Princeton) Matching and Weighting Methods Duke (January 8 9, 23) / 57

More information

Implementing Matching Estimators for. Average Treatment Effects in STATA

Implementing Matching Estimators for. Average Treatment Effects in STATA Implementing Matching Estimators for Average Treatment Effects in STATA Guido W. Imbens - Harvard University West Coast Stata Users Group meeting, Los Angeles October 26th, 2007 General Motivation Estimation

More information

Ordinary Least Squares Regression

Ordinary Least Squares Regression Ordinary Least Squares Regression Goals for this unit More on notation and terminology OLS scalar versus matrix derivation Some Preliminaries In this class we will be learning to analyze Cross Section

More information

New Developments in Econometrics Lecture 9: Stratified Sampling

New Developments in Econometrics Lecture 9: Stratified Sampling New Developments in Econometrics Lecture 9: Stratified Sampling Jeff Wooldridge Cemmap Lectures, UCL, June 2009 1. Overview of Stratified Sampling 2. Regression Analysis 3. Clustering and Stratification

More information

Questions and Answers on Unit Roots, Cointegration, VARs and VECMs

Questions and Answers on Unit Roots, Cointegration, VARs and VECMs Questions and Answers on Unit Roots, Cointegration, VARs and VECMs L. Magee Winter, 2012 1. Let ɛ t, t = 1,..., T be a series of independent draws from a N[0,1] distribution. Let w t, t = 1,..., T, be

More information

Specification Errors, Measurement Errors, Confounding

Specification Errors, Measurement Errors, Confounding Specification Errors, Measurement Errors, Confounding Kerby Shedden Department of Statistics, University of Michigan October 10, 2018 1 / 32 An unobserved covariate Suppose we have a data generating model

More information