서브메뉴
검색
Essays in Econometrics
Essays in Econometrics
상세정보
- 자료유형
- 학위논문 서양
- 최종처리일시
- 20250211151100
- ISBN
- 9798382757537
- DDC
- 310
- 저자명
- Rafi, Ahnaf.
- 서명/저자
- Essays in Econometrics
- 발행사항
- [Sl] : Northwestern University, 2024
- 발행사항
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- 형태사항
- 341 p
- 주기사항
- Source: Dissertations Abstracts International, Volume: 85-11, Section: B.
- 주기사항
- Advisor: Horowitz, Joel L.;Canay, Ivan A.
- 학위논문주기
- Thesis (Ph.D.)--Northwestern University, 2024.
- 초록/해제
- 요약This dissertation studies three distinct problems in econometrics. Chapter 1 studies inference on a class of functionals in a nonparametric variant of the random coefficients logit model. The random coefficients logit model is widely used in choice analysis, empirical industrial organization, and transport economics among other fields. Many objects of interest in this model can be represented as functionals of the distribution of random coefficients (RC/RCs), specifically averages. Some examples are: welfare measures, choice probabilities and their derivatives. Chapter 1 provides a nonparametric estimator of the RC distribution under which implied plug-in estimators of such averages are asymptotically normal. This is the first formal limiting distribution result for a nonparametric plug-in estimator of such functionals in the RC logit model. For the particular functionals considered here, this asymptotic normality occurs at the parametric n −1/2 rate. A consistent estimator of the variance of this limiting distribution is also provided. Together, these results make consistent tests of hypotheses and valid confidence intervals possible in the RC logit model when the distribution of RCs is estimated nonparametrically.Chapter 2 studies efficient estimation of the average treatment effect (ATE) in randomized experiments under a class of randomization procedures known as covariate adaptive randomization (CAR). Here, "efficient estimation" means construction of estimates that achieve the semiparametric efficiency bound (SEB) for the ATE. Experiments that use CAR are commonplace in applied economics and other fields. Under CAR, the experimenter first stratifies the sample according to observed baseline covariates and then assigns treatment randomly within these strata so as to achieve balance according to pre-specified stratum-specific target assignment proportions. We allow for the class of CAR procedures considered in Bugni et al. (2018, 2019). In Chapter 2, we first compute the SEB for estimating the ATE. The stratum-specific target proportions play the role of the propensity score conditional on all baseline covariates. The efficiency bound is a special case of the bound in Hahn (1998), but conditional on all baseline covariates, not just the stratum labels. Next, we show that this efficiency bound is achievable under the same (weak) conditions as those used to derive the bound. To do this, we construct an ATE estimator by combining the efficient influence function, a byproduct of the efficiency bound derivation, and a cross-fitted Nadaraya-Watson kernel regression estimator to form nonparametric regression adjustments.Chapter 3 (joint with Yong Cai) considers the issue of experiment design with the Neyman Allocation, which is used in many papers on experimental design. These papers typically assume that researchers have access to large pilot studies. This may be unrealistic. To understand the properties of the Neyman Allocation with small pilots, we study its behavior in an asymptotic framework that takes pilot size to be fixed even as the size of the main wave tends to infinity. Our analysis shows that the Neyman Allocation can lead to estimates of the ATE with higher asymptotic variance than with (non-adaptive) balanced randomization. In particular, this happens when the outcome variable is relatively homoskedastic with respect to treatment status or when it exhibits high kurtosis. We provide a series of empirical examples showing that such situations can arise in practice. Our results suggest that researchers with small pilots should not use the Neyman Allocation if they believe that outcomes are homoskedastic or heavy-tailed. Finally, we examine some potential methods for improving the finite sample performance of the FNA via simulations.
- 일반주제명
- Statistics
- 일반주제명
- Finance
- 키워드
- Econometrics
- 기타저자
- Northwestern University Economics
- 기본자료저록
- Dissertations Abstracts International. 85-11B.
- 전자적 위치 및 접속
- 로그인 후 원문을 볼 수 있습니다.
MARC
008250123s2024 us c eng d■001000017160682
■00520250211151100
■006m o d
■007cr#unu||||||||
■020 ▼a9798382757537
■035 ▼a(MiAaPQ)AAI31142825
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a310
■1001 ▼aRafi, Ahnaf.
■24510▼aEssays in Econometrics
■260 ▼a[Sl]▼bNorthwestern University▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a341 p
■500 ▼aSource: Dissertations Abstracts International, Volume: 85-11, Section: B.
■500 ▼aAdvisor: Horowitz, Joel L.;Canay, Ivan A.
■5021 ▼aThesis (Ph.D.)--Northwestern University, 2024.
■520 ▼aThis dissertation studies three distinct problems in econometrics. Chapter 1 studies inference on a class of functionals in a nonparametric variant of the random coefficients logit model. The random coefficients logit model is widely used in choice analysis, empirical industrial organization, and transport economics among other fields. Many objects of interest in this model can be represented as functionals of the distribution of random coefficients (RC/RCs), specifically averages. Some examples are: welfare measures, choice probabilities and their derivatives. Chapter 1 provides a nonparametric estimator of the RC distribution under which implied plug-in estimators of such averages are asymptotically normal. This is the first formal limiting distribution result for a nonparametric plug-in estimator of such functionals in the RC logit model. For the particular functionals considered here, this asymptotic normality occurs at the parametric n −1/2 rate. A consistent estimator of the variance of this limiting distribution is also provided. Together, these results make consistent tests of hypotheses and valid confidence intervals possible in the RC logit model when the distribution of RCs is estimated nonparametrically.Chapter 2 studies efficient estimation of the average treatment effect (ATE) in randomized experiments under a class of randomization procedures known as covariate adaptive randomization (CAR). Here, "efficient estimation" means construction of estimates that achieve the semiparametric efficiency bound (SEB) for the ATE. Experiments that use CAR are commonplace in applied economics and other fields. Under CAR, the experimenter first stratifies the sample according to observed baseline covariates and then assigns treatment randomly within these strata so as to achieve balance according to pre-specified stratum-specific target assignment proportions. We allow for the class of CAR procedures considered in Bugni et al. (2018, 2019). In Chapter 2, we first compute the SEB for estimating the ATE. The stratum-specific target proportions play the role of the propensity score conditional on all baseline covariates. The efficiency bound is a special case of the bound in Hahn (1998), but conditional on all baseline covariates, not just the stratum labels. Next, we show that this efficiency bound is achievable under the same (weak) conditions as those used to derive the bound. To do this, we construct an ATE estimator by combining the efficient influence function, a byproduct of the efficiency bound derivation, and a cross-fitted Nadaraya-Watson kernel regression estimator to form nonparametric regression adjustments.Chapter 3 (joint with Yong Cai) considers the issue of experiment design with the Neyman Allocation, which is used in many papers on experimental design. These papers typically assume that researchers have access to large pilot studies. This may be unrealistic. To understand the properties of the Neyman Allocation with small pilots, we study its behavior in an asymptotic framework that takes pilot size to be fixed even as the size of the main wave tends to infinity. Our analysis shows that the Neyman Allocation can lead to estimates of the ATE with higher asymptotic variance than with (non-adaptive) balanced randomization. In particular, this happens when the outcome variable is relatively homoskedastic with respect to treatment status or when it exhibits high kurtosis. We provide a series of empirical examples showing that such situations can arise in practice. Our results suggest that researchers with small pilots should not use the Neyman Allocation if they believe that outcomes are homoskedastic or heavy-tailed. Finally, we examine some potential methods for improving the finite sample performance of the FNA via simulations.
■590 ▼aSchool code: 0163.
■650 4▼aStatistics
■650 4▼aFinance
■653 ▼aRandom coefficients
■653 ▼aEconometrics
■653 ▼aAverage treatment effect
■653 ▼aCovariate adaptive randomization
■690 ▼a0501
■690 ▼a0508
■690 ▼a0463
■71020▼aNorthwestern University▼bEconomics.
■7730 ▼tDissertations Abstracts International▼g85-11B.
■790 ▼a0163
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17160682▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.


