Download or read book Essays on Causal Inference and Econometrics written by Haitian Xie and published by . This book was released on 2023 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: This dissertation is a collection of three essays on the econometric analysis of causal inference methods. Chapter 1 examines the identification and estimation of the structural function in fuzzy RD designs with a continuous treatment variable. We show that the nonlinear and nonseparable structural function can be nonparametrically identified at the RD cutoff under shape restrictions, including monotonicity and smoothness conditions. Based on the nonparametric identification equation, we propose a three-step semiparametric estimation procedure and establish the asymptotic normality of the estimator. The semiparametric estimator achieves the same convergence rate as in the case of a binary treatment variable. As an application of the method, we estimate the causal effect of sleep time on health status by using the discontinuity in natural light timing at time zone boundaries. Chapter 2 examines the local linear regression (LLR) estimate of the conditional distribution function F(y|x). We derive three uniform convergence results: the uniform bias expansion, the uniform convergence rate, and the uniform asymptotic linear representation. The uniformity in the above results is with respect to both x and y and therefore has not previously been addressed in the literature on local polynomial regression. Such uniform convergence results are especially useful when the conditional distribution estimator is the first stage of a semiparametric estimator. Chapter 3 studies the estimation of causal parameters in the generalized local average treatment effect model, a generalization of the classical LATE model encompassing multi-valued treatment and instrument. We derive the efficient influence function (EIF) and the semiparametric efficiency bound for two types of parameters: local average structural function (LASF) and local average structural function for the treated (LASF-T). The moment condition generated by the EIF satisfies two robustness properties: double robustness and Neyman orthogonality. Based on the robust moment condition, we propose the double/debiased machine learning (DML) estimators for LASF and LASF-T. We also propose null-restricted inference methods that are robust against weak identification issues. As an empirical application, we study the effects across different sources of health insurance by applying the developed methods to the Oregon Health Insurance Experiment.

Download or read book Essays on Causal Inference in Econometrics written by Hugo Bodory and published by . This book was released on 2016 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: This doctoral thesis consists of four chapters. Each of the studies builds on the concept of causal inference. Two papers are empirical applications that analyze the effects of welfare dependency on health and health-related behavior. The remaining papers are methodological contributions to the literature on treatment effects, which focus on the introduction and evaluation of inference methods. The first chapter investigates whether welfare dependency has an impact on individual health and health-related behavior. The empirical analysis uses panel survey data to study health-related effects of the major German welfare program Hartz IV. Using a sample of individuals initially on welfare, the paper compares the health outcomes of two groups: those who remain on welfare and those who get off welfare. The findings show that welfare dependency can be detrimental to the health of individuals, as well as to their sports-related behavior. The second chapter conducts a mediation analysis to identify potential channels that can influence the health conditions of welfare recipients. The study uses a semi-parametric estimation method especially adapted to this mediation analysis to compute the effects on health. Evidence suggests that employment enhances the health of males and older individuals when getting off welfare. In contrast, health improvements for females cannot be attributed to employment but to the direct (or residual) effect of leaving welfare. The health of younger individuals is not affected by welfare dependency. The third chapter investigates the finite sample properties of a range of inference methods for treatment effect estimators. The simulations, based on empirical data, use both asymptotic approximations of analytical variances and bootstrap methods to compute confidence intervals and p-values. The results suggest that, in general, the bootstrap approaches outperform the analytical variance approximations in terms of s.

Download or read book Essays on Econometrics Causal Inference and Machine Learning written by Rahul Singh (Econometrician) and published by . This book was released on 2023 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: The traditional tools of econometrics may be inadequate for modern data sets, for example the 2020 US Census, which will be deliberately corrupted by the Census Bureau in the interest of privacy. Meanwhile, the modern tools of machine learning may be inadequate for the traditional goals of policy evaluation, which are to measure cause and effect and to assess statistical significance. In this dissertation, I develop tools for flexible causal inference, weaving machine learning into econometrics and solving unique problems that arise at their intersection. Specifically, I work in three domains at the intersection between econometrics and machine learning: (Chapter 1) causal inference with privacy protected data, (Chapter 2) rigorous statistical guarantees for machine learning, and (Chapter 3) simple algorithms for complex causal problems. JEL: C81,C45,C26.

Download or read book Essays in Econometrics written by Dmitry Arkhangelskiy and published by . This book was released on 2018 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: In this dissertation, I propose novel approaches to causal inference in the settings characterized by an explicit clustering structure. I study different aspects of this problem, considering settings with few large clusters as well as with many small clusters. The dissertation consists of two essays. The first essay proposes a new model for causal inference in the settings with few large clusters and cluster-level treatment assignment. The second essay studies causal inference questions in the settings with many clusters of moderate size and individual-level treatment assignment. In the first essay, I construct a nonlinear model for causal inference in the empirical settings where researchers observe individual-level data for few large clusters over at least two time periods. It allows for identification (sometimes partial) of the counterfactual distribution, in particular, identifying average treatment effects and quantile treatment effects. The model is flexible enough to handle multiple outcome variables, multidimensional heterogeneity, and multiple clusters. It applies to the settings where the new policy is introduced in some of the clusters, and a researcher additionally has information about the pretreatment periods. I argue that in such environments we need to deal with two different sources of bias: selection and technological. In my model, I employ standard methods of causal inference to address the selection problem and use pretreatment information to eliminate the technological bias. In case of one-dimensional heterogeneity, identification is achieved under natural monotonicity assumptions. The situation is considerably more complicated in case of multidimensional heterogeneity where I propose three different approaches to identification using results from transportation theory. The second essay is co-authored with Guido Imbens. We develop a new estimator for the average treatment effect in the observational studies with unobserved cluster-level heterogeneity. We show that under particular assumptions on the sampling scheme the unobserved confounders can be integrated out conditioning on the empirical distribution of covariates and policy variable within the cluster. To make this result practical we impose a particular exponential family structure that implies that a low-dimensional sufficient statistic can summarize the empirical distribution. Then we use modern causal inference methods to construct a novel doubly robust estimator. The proposed estimator uses the estimated propensity score to adjust the familiar fixed effect estimator.

Download or read book Essays on Applied Econometrics and Causal Inference Applications to the Analysis of the Tax Multiplier and to the Evaluation of Online Lending Market written by Wei Xu and published by . This book was released on 2018 with total page 192 pages. Available in PDF, EPUB and Kindle. Book excerpt: The dissertation consists of three chapters, with emphasis on analyzing macro- and micro-level data and applying econometric techniques so as to measure treatment effects and draw a causal inference.

Download or read book Essays on the Econometrics of Causal Inference Resampling and Spatial Dependence written by Marinho Angelo Bertanha and published by . This book was released on 2015 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: This thesis is a collection of four papers corresponding to all the research in econometrics that I have done during my graduate studies at Stanford. The first and second chapters study causal inference in regression discontinuity designs. In recent years, numerous studies have employed regression discontinuity designs with many cutoffs assigning individuals to heterogeneous treatments. A common practice is to normalize all of the cutoffs to zero and estimate only one effect. This procedure identifies the average of local treatment effects weighted by the observed relative density of individuals at the existing cutoffs. However, researchers often want to make inferences on more meaningful average treatment effects (ATE) computed over general counterfactual distributions of individuals rather than simply the observed distribution of individuals local to existing cutoffs. In the first chapter, we propose a root-n consistent and asymptotically normal estimator for such ATEs when heterogeneity follows a non-parametric smooth function of cutoff characteristics. In the case of parametric heterogeneity, observations are optimally combined to minimize the mean squared error of the ATE estimator. Inference results are also provided for the fuzzy regression discontinuity case, where the parametric heterogeneity assumption yields identification of treatment effects on individuals who comply with at least one of the multiple treatments. In the second chapter, we focus on Fuzzy Regression Discontinuity (FRD) designs with one cutoff. Many empirical studies use FRD designs to identify treatment effects when the receipt of treatment is potentially correlated to outcomes. Existing FRD methods identify the local average treatment effect (LATE) on the subpopulation of compliers with values of the forcing variable that are equal to the threshold. In the second chapter, we develop methods that assess the plausibility of generalizing LATE to subpopulations other than compliers, and to subpopulations other than those with forcing variable equal to the threshold. Specifically, we focus on testing the equality of the distributions of potential outcomes for treated compliers and always-takers, and for non-treated compliers and never-takers. We show that equality of these pairs of distributions implies that the expected outcome conditional on the forcing variable and the treatment status is continuous in the forcing variable at the threshold, for each of the two treatment regimes. As a matter of routine, we recommend that researchers present graphs with estimates of these two conditional expectations in addition to graphs with estimates of the expected outcome conditional on the forcing variable alone. We illustrate our methods using data on the academic performance of students attending the summer school program in two large school districts in the US. In the third chapter, we propose a fast resample method for two step nonlinear parametric and semiparametric models. Our resample method is faster than standard methods because it does not require recomputation of the second stage estimator during each resample iteration. The fast resample method directly exploits the score function representations computed on each bootstrap sample, thereby reducing computational time considerably. This method is used to approximate the limit distribution of parametric and semiparametric estimators, possibly simulation based, that admit an asymptotic linear representation. Monte Carlo experiments demonstrate the desirable performance and vast improvement in the numerical speed of the fast bootstrap method. Finally, the fourth chapter studies the effects of spatially correlated data on count data regressions. Count data regressions are an important tool for empirical analyses ranging from analyses of patent counts to measures of health and unemployment. Along with negative binomial, Poisson panel regressions are a preferred method of analysis because the Poisson conditional fixed effects maximum likelihood estimator (PCFE) and its sandwich variance estimator are consistent even if the data are not Poisson-distributed, or if the data are correlated over time. Analyses of counts may however also be affected by correlation in the cross-section. For example, patent counts or publications may increase across related research fields in response to common shocks. The fourth chapter shows that the PCFE and its sandwich variance estimator are consistent in the presence of such dependence in the cross-section - as long as spatial dependence is time-invariant. We develop a test for time-invariant spatial dependence and provide code in STATA and MATLAB to implement the test.

Download or read book Essays in High dimensional Econometrics written by and published by . This book was released on 2014 with total page 168 pages. Available in PDF, EPUB and Kindle. Book excerpt: The dissertation contains three papers on causal inference in econometrics.

Download or read book Essays in Econometrics and Industrial Organization written by Nikolay Doudchenko and published by . This book was released on 2018 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: In my dissertation I study econometric methods of causal inference with a particular focus on the use of prediction methods developed by researchers in the fields of statistical learning, machine learning, and pattern recognition. I'm also interested in the application of these methods as well as the more traditional ones to answer relevant policy questions. Chapter 1 (joint with Guido Imbens) considers the synthetic control method developed by Abadie, Diamond, Gardeazabal, and Hainmueller in several influential papers. The method is designed for estimating the effect of a treatment, in the presence of a single treated unit and a number of control units, with pre-treatment outcomes observed for all units. The method constructs a set of weights such that selected covariates and pre-treatment outcomes of the treated unit are approximately matched by a weighted average of the control units (the synthetic control). The weights are restricted to be nonnegative and sum to one. These restrictions are important partly because they make it easier for the procedure to obtain unique weights even when the number of lagged outcomes is modest relative to the number of control units, a common setting in applications. In the chapter we propose a generalization of the synthetic control procedure that allows the weights to be negative, and their sum to differ from one, and that allows for a permanent additive difference between the treated unit and the controls, similar to the difference-in-difference procedures. The weights directly minimize the distance between the lagged outcomes for the treated and the control units, using elastic net regularization to deal with a potentially large number of possible control units. In Chapter 2 (joint with Ali Yurukoglu) we quantify how bargaining power derived from firm size affects the analysis of downstream mergers and the profitability of downstream entry in the multichannel television industry. We estimate an empirical model of the industry which features negotiations between the upstream content producers and the downstream distributors of varying size. We estimate that large distributors like Comcast are able to negotiate about 25% lower content fees than smaller downstream firms such as Cablevision. We evaluate the short-run welfare effects of several recently reviewed mergers taking into account the size effects in negotiations. We also assess the degree to which size based bargaining power creates contracts which are a barrier to entry for new distributors.

Download or read book Essays in Causal Inference with Panel Data written by Timo Schenk and published by . This book was released on 2024 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: "This dissertation discusses and advances the econometric methods in the field of causal inference with panel data. In particular, it improves difference-in-differences methods in multiple aspects. Researchers in all fields of economics apply these methods to answer important questions, such as “what is the effect of a labor market interventions on earnings?”, “to what extent do changes in the school leaving age affect study choices?” or “by how much have the policy instruments of the clean-air-act reduced emissions of greenhouse gases?”."--

Download or read book Essays on Causal Inference in Economics written by Elias Moor and published by . This book was released on 2020 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt:

Download or read book Three Essays on Causal Inference for Marketing Applications written by Ashutosh Charudatta Bhave and published by . This book was released on 2022 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: In my dissertation consisting of three research projects, I focus on solving problems which deal with reliably estimating the impact of a change in policy in quasi-experimental setup. I utilize cutting edge methods in econometrics and machine learning to quantify causal effects of policy changes, understand the mechanism behind the effect and most importantly highlight the implications for the managers and policy makers. My first research paper, “A Study of the Effects of Legalization of Recreational Marijuana on Sales of Cigarettes” attempts to establish a causal link between the legalization of recreational marijuana and the sales of cigarettes in retail stores. Recreational marijuana legalization (RML) has been on the rise in the recent years and many arguments have been put forth to support or counter this move. We explore the possibility of RML impacting cigarette consumption. This is important for understanding the impact on health care expenditures related to smoking, which is about $330 billion in the US. Our results show that in states that have passed RML, there is a 7% increase in cigarette sales. This is an important finding since it reverses a decline in cigarette sales in recent years. Therefore, we conclude that states should exercise caution while considering legalization of recreational use of marijuana. My second project, “Effects of Social Media Fights and New Product Launches in the Fast Food Industry” examines the effects of engaging in ‘Twitter feuds’ with competition during new product launches. We propose a viable mechanism that explains how seemingly harmless banter of social media could have unforeseen impact on a firm’s business. Through empirical evidence from recent incidents, we show that Twitter activity has a spillover into traditional media which leads to surge in online search. Online search activity is followed by the offline sales as documented in literature as well as evidenced from our unique foot traffic data. Next, we document the long-term effects of this menu innovation in causal framework, well beyond the initial frenzy, with a novel synthetic difference-in-differences (SDID) method proposed by Arkhangelsky et al. (2021). Results show that the launch led to a 30% increase in store visits up to six months after the launch. Overall, these findings underscore the importance of savvy social media presence especially during a product launch- which could be driver for peaked interest leading to impact on overall business. The flip side for competitors is that initiating seemingly harmless banter, unlike in the offline setting, could end up providing free publicity to one’s rivals. Overall, we highlight the enormous potential of social media to affect business and advise caution to brand managers before engaging in any activity. My third project “A study of wear out and heterogeneous effects of unlimited shipping program on customer engagement in the online retail industry” we study effects of a variation of free shipping promotion in the online retail industry. Free shipping promotions have become popular among online retailers. Most online shoppers expect deliveries without additional costs and cite it as a primary concern while shopping online. Many online retailers across industries have implemented long term free shipping programs on all purchases with fixed annual fees. In this paper, we analyze benefits associated with such programs for the retailers and also shed light on the potential pitfalls, using data from a leading online retailer in the UK. Our results indicate that that there is a significant decay in customer spending after initial days and the effects wear out completely short way through the promotion period. Moreover, changes in purchase behavior (significantly lower basket size after enrolling for free shipping) could hurt the retailer. Thus, online retailers should be cautious when offering long term free shipping promotion. In the next part of the paper, we use pre-promotion engagement as a moderating factor to capture heterogeneous effects of free shipping programs across customers, using Honest Causal Forests approach. Our results show that free shipping promotions work better (higher revenues, smaller drop in basket size) for customers with relatively lower engagement with the retailer in the prepromotion period. Online retailers could use these findings to devise their targeting strategy for free shipping promotions.

Download or read book Causal Inference in Econometrics written by Van-Nam Huynh and published by Springer. This book was released on 2015-12-28 with total page 626 pages. Available in PDF, EPUB and Kindle. Book excerpt: This book is devoted to the analysis of causal inference which is one of the most difficult tasks in data analysis: when two phenomena are observed to be related, it is often difficult to decide whether one of them causally influences the other one, or whether these two phenomena have a common cause. This analysis is the main focus of this volume. To get a good understanding of the causal inference, it is important to have models of economic phenomena which are as accurate as possible. Because of this need, this volume also contains papers that use non-traditional economic models, such as fuzzy models and models obtained by using neural networks and data mining techniques. It also contains papers that apply different econometric models to analyze real-life economic dependencies.

Download or read book Essays on Impact Evaluation in Labor and Development Economics written by Garret Smyth Christensen and published by . This book was released on 2011 with total page 212 pages. Available in PDF, EPUB and Kindle. Book excerpt: This dissertation studies examples of applied econometrics for causal inference in labor and development economics. One of the fundamental problems in applied fields of economics is causal inference. Merely observing that event B occurred after event A is not enough to claim that A caused B. The field of economics, and the social sciences in general, are limited by ethics and practicality in their ability to conduct randomized field experiments, the gold standard for causality in other fields. Several statistical methods have been devised to obtain causal estimates from "natural" or "quasi" experimental settings--settings where plausibly exogenous variation in a treatment effect of interest can be found and exploited to produce an unbiased estimate of causal effects. Some of these methods include panel data with fixed effects, nearest-neighbor matching, and regression discontinuity. This dissertation explores applications of these econometric methods, as well as an actual randomized controlled trial, in issues of labor and development economics. The first chapter uses panel data, and causal estimates are identified using a series of fixed effects to control for unmeasurable characteristics that could be correlated with both dependent and independent variables. The subject matter is the recruiting task of the United States military, which is the largest employer in the nation and spends over $4 billion each year to recruit roughly 200,000 new soldiers to maintain its troop levels. This recruiting task has become more expensive since the beginning of the wars in Iraq and Afghanistan. I use a detailed new dataset of all US military applicants over several recent years and find that deaths in Iraq of US soldiers had a significant deterrent effect on recruiting in the home county of the soldiers who were killed. The deterrent effect of local deaths is significantly larger than the deterrent from a death from outside the county. The deterrent exhibits significant heterogeneity across characteristics of deaths, recruits, and locations. Deaths from Iraq decrease recruiting, while deaths from Afghanistan actually increase recruiting. Recruits with higher test scores are more deterred by deaths, and the deterrent is larger and more negative in less populous and more racially diverse counties, but is significantly smaller and in many cases even positive in counties that voted for George W. Bush in the 2004 presidential election. The findings provide strong evidence that recruits are over-emphasizing local information and have war-specific tastes and preferences that makes enlistment decisions more complicated than a full-information utility-maximization model of risk and monetary compensation would predict. The second chapter uses nearest-neighbor matching techniques to look at performance of Major League Baseball players after they win awards in order to shed light on the more general question of how rational agents perform after they have been rewarded for good behavior up to that point. Comparing individual player's performance after winning major awards to their performance before winning shows that although players do perform significantly better in the year in which they win the award, performance after the award is generally indistinguishable from pre-award performance. Matching methods based on both baseball writer voting and performance statistics also indicate the likely absence of any sort of "curse" from winning awards for the winners themselves, their teams, and their teammates. The third chapter, which is co-authored work with Michael Kremer and Edward Miguel, uses data from a randomized controlled trial, the Girls Scholarship Program (GSP), as well as the Kenya Life Panel Survey (KLPS) to conduct three types of analysis of bursary programs. We evaluate the effect of different targeting rules for secondary school scholarships, we estimate the impact of attending a primary school that took part in a scholarship program, and we estimate the effect of winning a scholarship from the program. Giving scholarships based on KCPE alone would lead to under representation of children whose parents have no secondary education and girls relative to their proportion of the population. Distributing the scholarships to the top students in each school as opposed to each district does little to alleviate this discrepancy. Analysis of the medium-run impacts of the Girls Scholarship Program, gave largely inconclusive but suggestive evidence that there were moderate benefits from attending a scholarship program school on the order of one half of the benefits observed in the original study held immediately after the scholarship program. The evidence indicates that scholarship winners did not benefit greatly from the award itself.

Download or read book Identification and Inference for Econometric Models written by Donald W. K. Andrews and published by Cambridge University Press. This book was released on 2005-07-04 with total page 589 pages. Available in PDF, EPUB and Kindle. Book excerpt: This 2005 volume contains the papers presented in honor of the lifelong achievements of Thomas J. Rothenberg on the occasion of his retirement. The authors of the chapters include many of the leading econometricians of our day, and the chapters address topics of current research significance in econometric theory. The chapters cover four themes: identification and efficient estimation in econometrics, asymptotic approximations to the distributions of econometric estimators and tests, inference involving potentially nonstationary time series, such as processes that might have a unit autoregressive root, and nonparametric and semiparametric inference. Several of the chapters provide overviews and treatments of basic conceptual issues, while others advance our understanding of the properties of existing econometric procedures and/or propose others. Specific topics include identification in nonlinear models, inference with weak instruments, tests for nonstationary in time series and panel data, generalized empirical likelihood estimation, and the bootstrap.

Download or read book Essays on Causal Inference in Randomized Experiments written by Winston Lin and published by . This book was released on 2013 with total page 186 pages. Available in PDF, EPUB and Kindle. Book excerpt: This dissertation explores methodological topics in the analysis of randomized experiments, with a focus on weakening the assumptions of conventional models. Chapter 1 gives an overview of the dissertation, emphasizing connections with other areas of statistics (such as survey sampling) and other fields (such as econometrics and psychometrics). Chapter 2 reexamines Freedman's critique of ordinary least squares regression adjustment in randomized experiments. Using Neyman's model for randomization inference, Freedman argued that adjustment can lead to worsened asymptotic precision, invalid measures of precision, and small-sample bias. This chapter shows that in sufficiently large samples, those problems are minor or easily fixed. OLS adjustment cannot hurt asymptotic precision when a full set of treatment-covariate interactions is included. Asymptotically valid confidence intervals can be constructed with the Huber-White sandwich standard error estimator. Checks on the asymptotic approximations are illustrated with data from a randomized evaluation of strategies to improve college students' achievement. The strongest reasons to support Freedman's preference for unadjusted estimates are transparency and the dangers of specification search. Chapter 3 extends the discussion and analysis of the small-sample bias of OLS adjustment. The leading term in the bias of adjustment for multiple covariates is derived and can be estimated empirically, as was done in Chapter 2 for the single-covariate case. Possible implications for choosing a regression specification are discussed. Chapter 4 explores and modifies an approach suggested by Rosenbaum for analysis of treatment effects when the outcome is censored by death. The chapter is motivated by a randomized trial that studied the effects of an intensive care unit staffing intervention on length of stay in the ICU. The proposed approach estimates effects on the distribution of a composite outcome measure based on ICU mortality and survivors' length of stay, addressing concerns about selection bias by comparing the entire treatment group with the entire control group. Strengths and weaknesses of possible primary significance tests (including the Wilcoxon-Mann-Whitney rank sum test and a heteroskedasticity-robust variant due to Brunner and Munzel) are discussed and illustrated.

Download or read book Essays in Public Economics and Applied Econometrics written by Soo Kyo Jeong and published by . This book was released on 2021 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: This dissertation consists of two chapters of in-depth empirical analyses in the domain of public economics, and one chapter of causal inference methodology regarding distributional robustness. The first chapter is titled ``The Effect of Air Pollution on Academic Performances: Evidence from South Korea, " and investigates the causal effect of air pollution on academic performances. While the effect of air pollution on health has been extensively studied, little is known about its effect on education, especially in a causal context. Here, I exploit the unique geography of Korea and a meteorological phenomenon called Asian Dust Storm (ADS) to get exogenous shocks of air pollution carried by the wind from China. Using two stage least squares regression, I find that an increase in particulate matter (PM10) leads to an increase in the share of students who underperform, while its effect on the share of students who overperform is not different from zero. I find similar results for elementary and middle school test outcomes, and find that air pollution disproportionately affects the types of schools associated with low socioeconomic status. Looking at both short term and long term effect of air pollutants, I find that air pollution has both acute and cumulative effect on the academic performances of the students. I explore health as a mechanism through which air pollution affects academic outcomes, and find that the most detrimental effect comes from the most harmful to health pollutants--PM10 and ozone--and do not find any evidence of preemptive absenteeism or mobilization due to air pollution on the school level. The second chapter, co-authored with Mark Duggan, Irena Dushi, and Gina Li, is titled ``The Effects of Changes in Social Security's Delayed Retirement Credit: Evidence from Administrative Data." The delayed retirement credit (DRC) increases monthly OASI (Old Age and Survivors Insurance) benefits for primary beneficiaries who claim after their full retirement age (FRA). For many years, the DRC was set at 3.0 percent per year (0.25 percent monthly). The 1983 amendments to Social Security more than doubled this actuarial adjustment to 8.0 percent per year. These changes were phased in gradually, so that those born in 1924 or earlier retained a 3.0 percent DRC while those born in 1943 or later had an 8.0 percent DRC. In this paper, we use administrative data from the Social Security Administration (SSA) to estimate the effect of this policy change on individual claiming behavior. We focus on the first half of the DRC increase (from 3.0 to 5.5 percent) given changes in other SSA policies that coincided with the later increases. Our findings demonstrate that the increase in the DRC led to a significant increase in delayed claiming of social security benefits and strongly suggest that the effects were larger for those with higher lifetime incomes, who would have a greater financial incentive to delay given their longer life expectancies. The third chapter, co-authored with Hongseok Namkoong, is titled `Robust Causal Inference Under Covariate Shift via Worst-Case Subpopulation Treatment Effects." In this chapter, we propose the worst-case treatment effect (WTE) across all subpopulations of a given size, a conservative notion of topline treatment effect. Compared to the average treatment effect (ATE), whose validity relies on the covariate distribution of collected data, WTE is robust to unanticipated covariate shifts, and positive findings guarantee uniformly valid treatment effects over subpopulations. We develop a semiparametrically efficient estimator for the WTE, leveraging machine learning-based estimates of the heterogeneous treatment effect and propensity score. By virtue of satisfying a key (Neyman) orthogonality property, our estimator enjoys central limit behavior--oracle rates with true nuisance parameters--even when estimates of nuisance parameters converge at slower rates. For both randomized trials and observational studies, we establish a semiparametric efficiency bound, proving that our estimator achieves the optimal asymptotic variance. On real datasets where robustness to covariate shift is of core concern, we illustrate the non-robustness of ATE under even mild distributional shift, and demonstrate that the WTE guards against brittle findings that are invalidated by unanticipated covariate shifts.

Download or read book Essays in Causal Inference written by Yoshiyasu Rai and published by . This book was released on 2019 with total page 112 pages. Available in PDF, EPUB and Kindle. Book excerpt: In Chapter 1, I study the statistical inference problem for treatment assignment policies. In typical applications, individuals with different characteristics are expected to differ in their responses to treatment. As a result, treatment assignment policies that allocate treatment based on individuals' observed characteristics can have a significant influence on outcomes and welfare. A growing literature proposes various approaches to estimating the welfare-optimzing treatment assignment policy. I develop a method for assessing the precision of estimated optimal policies. In particular, for the welfare used by \cite{KT:18} to propose estimated assignment policy, my method constructs (i) a confidence set of policies that contains the optimal policy, which maximizes the average social welfare among all the feasible policies with prespecified level and (ii) a confidence interval for the maximized welfare. A simulation study indicates that the proposed methods work reasonably well with modest sample size. I apply the method to experimental data from the National Job Training Partnership Act study. In Chapter 2, I derive the large sample properties of $M$th nearest neighbor propensity score matching estimator with a potentially misspecified propensity score model. By using the local misspecification framework, I formalize the bias/variance trade-off with respect to the choice of propensity score estimator and propose a model selection criterion that aims to minimize the estimation error. Finally, in Chapter 3 (co-authored with Taisuke Otsu), we propose asymptotically valid inference methods for matching estimators based on the weighted bootstrap. The key is to construct bootstrap counterparts by resampling based on certain linear forms of the estimators. Our weighted bootstrap is applicable for the matching estimators of both the average treatment effect and its counterpart for the treated population. Also, by incorporating a bias correction method in \cite{AI:11}, our method can be asymptotically valid even for matching based on a vector of covariates. A simulation study indicates that the weighted bootstrap method is favorably comparable with the asymptotic normal approximation by \cite{AI:06}. As an empirical illustration, we apply the proposed method to the National Supported Work data.