東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

Essays on applied semiparametric eco...

Rau Binder, Tomas Andres.

FindBook

Google Book

Amazon

博客來

Essays on applied semiparametric econometrics.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	Essays on applied semiparametric econometrics./
作者:	Rau Binder, Tomas Andres.
面頁冊數:	81 p.
附註:	Adviser: David Card.
Contained By:	Dissertation Abstracts International68-08A.
標題:	Economics, Labor. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275575
ISBN:	9780549168775

Essays on applied semiparametric econometrics.
Rau Binder, Tomas Andres.

Essays on applied semiparametric econometrics. - 81 p.

Adviser: David Card.

Thesis (Ph.D.)--University of California, Berkeley, 2007.

This dissertation consists of two essays. In the first essay I propose a new class of flexible microeconometric models that incorporates endogenous regressors and sorting. Existing models, including Garen's (1984) correlated random coefficients model, and Newey, Powell, and Vella's (1999) fully-flexible model with an additive error, can be derived from a structural equation with unobserved heterogeneity by imposing homogeneity and constancy assumptions on the first and second derivatives. I consider a less restrictive model that imposes homogeneity assumptions on the second partial derivative of the structural equation. Assuming the existence of suitable instrumental variables, the model can be estimated using a generalized control function approach. I consider an application to the estimation of the returns to education in Chile, exploiting variation across regions and cohorts in educational infrastructure and compulsory schooling laws. Using penalized spline functions to approximate the components of the average structural response function, I find that the local average returns to schooling are typically under-estimated by flexible models that ignore the endogeneity of schooling. I also find limited evidence of comparative advantage bias in the returns to certain levels of education.

ISBN: 9780549168775Subjects--Topical Terms:

1019135
Economics, Labor.

Essays on applied semiparametric econometrics.
LDR:03590nam 2200289 a 45 001 957287
005 20110630
008 110630s2007 ||||||||||||||||| ||eng d
020 $a 9780549168775
035 $a (UMI)AAI3275575
035 $a AAI3275575
040 $a UMI $c UMI
100 1 $a Rau Binder, Tomas Andres. $3 1280642
245 1 0 $a Essays on applied semiparametric econometrics.
300 $a 81 p.
500 $a Adviser: David Card.
500 $a Source: Dissertation Abstracts International, Volume: 68-08, Section: A, page: 3531.
502 $a Thesis (Ph.D.)--University of California, Berkeley, 2007.
520 $a This dissertation consists of two essays. In the first essay I propose a new class of flexible microeconometric models that incorporates endogenous regressors and sorting. Existing models, including Garen's (1984) correlated random coefficients model, and Newey, Powell, and Vella's (1999) fully-flexible model with an additive error, can be derived from a structural equation with unobserved heterogeneity by imposing homogeneity and constancy assumptions on the first and second derivatives. I consider a less restrictive model that imposes homogeneity assumptions on the second partial derivative of the structural equation. Assuming the existence of suitable instrumental variables, the model can be estimated using a generalized control function approach. I consider an application to the estimation of the returns to education in Chile, exploiting variation across regions and cohorts in educational infrastructure and compulsory schooling laws. Using penalized spline functions to approximate the components of the average structural response function, I find that the local average returns to schooling are typically under-estimated by flexible models that ignore the endogeneity of schooling. I also find limited evidence of comparative advantage bias in the returns to certain levels of education.
520 $a The second essay provides a new methodology for estimating the average treatment effect in a regression discontinuity (RD) design using penalized regression splines. In a RD design the treatment assignment or its probability is a discontinuous function of an observed variable (the index) at a known threshold value. The parameter of interest is the discontinuity in the conditional expectation of the outcome variable on the index generated by the assignment rule. Unweighted and locally weighted polynomials have been proposed for point estimation of the treatment effect in this context, but there has been little discussion about optimal smoothing, inference, and interval estimation. Given that semiparametric estimates are typically biased in small samples, interval estimation is an appealing way of overcoming this problem for inference purposes. I propose a penalized spline approach to estimate the model and the use of generalized cross validation to choose the smoothing parameter. I exploit the Bayesian interpretation of penalized regressions to obtain the standard errors from the posterior variance-covariance matrix. Montecarlo experiments show that confidence intervals based on the posterior variance-covariance matrix have close to nominal realized coverage probabilities and outperform intervals based on standard errors from usual estimators, such as local linear and local cubic polynomials.
590 $a School code: 0028.
650 4 $a Economics, Labor. $3 1019135
650 4 $a Economics, Theory. $3 1017575
690 $a 0510
690 $a 0511
710 2 $a University of California, Berkeley. $3 687832
773 0 $t Dissertation Abstracts International $g 68-08A.
790 $a 0028
790 1 0 $a Card, David, $e advisor
791 $a Ph.D.
792 $a 2007
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3275575