Muutke küpsiste eelistusi

E-raamat: Applied Nonparametric Econometrics

(University of Miami), (University of Alabama)
  • Formaat: PDF+DRM
  • Ilmumisaeg: 12-Jan-2015
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781316055939
Teised raamatud teemal:
  • Formaat - PDF+DRM
  • Hind: 49,39 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
  • Formaat: PDF+DRM
  • Ilmumisaeg: 12-Jan-2015
  • Kirjastus: Cambridge University Press
  • Keel: eng
  • ISBN-13: 9781316055939
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

"The majority of empirical research in economics ignores the potential benefits of nonparametric methods, while the majority of advances in nonparametric theory ignores the problems faced in applied econometrics. This book helps bridge this gap between applied economists and theoretical nonparametric econometricians. It discusses in depth, and in terms that someone with only one year of graduate econometrics can understand, basic to advanced nonparametric methods. The analysis starts with density estimation and motivates the procedures through methods that should be familiar to the reader. It then moves on to kernel regression, estimation with discrete data, and advanced methods such as estimation with panel data and instrumental variables models. The book pays close attention to the issues that arise with programming, computing speed, and application. In each chapter, the methods discussed are applied to actual data, paying attention to presentation of results and potential pitfalls"--

Arvustused

'A clear and thorough treatment of nonparametric and semiparametric econometrics. The text will be valuable to empirical researchers, who can expand their methodological toolkits without resorting to difficult journal articles. Even advanced topics, such as nonparametric instrumental variables and nonparametric models with panel data, are treated at an accessible level.' Jeffrey M. Wooldridge, Michigan State University 'Taking theory to data is difficult for most students, but this book provides substantial help by providing cogent explanations of practical considerations, including how well methods that work 'in theory' might be expected to work with real data in the quantities that researchers might have available.' Paul W. Wilson, Clemson University 'Daniel Henderson and Chris Parmeter have provided a modern survey of nonparametric econometrics. Newcomers will enjoy their applications-oriented introduction to this growing field. Theorists will find a compact survey of the most important foundations. Researchers of all sorts will want to add this valuable resource to their libraries.' William Greene, Stern School of Business, New York University 'This well-written textbook represents a rigorous yet accessible introduction to nonparametric methods, one that makes clear the importance of these techniques for empirical research. Henderson and Parmeter have performed a valuable service for students throughout the social sciences.' Steven N. Durlauf, University of Wisconsin 'Nonparametric econometric methods have by now become quite common in applied research, yet, as in almost all areas of research, theory precedes practice. The current hands-on approach of the book comes to fill the gap and offer the applied researcher a manual of how to properly use these methods without compromising rigor. It will complement other more theoretical books on the subject and as such it will prove very useful to many practitioners and students alike.' Thanasis Stengos, University of Guelph 'The authors advertise right at the beginning that this book was written to help bridge the gap between applied economists and theoretical nonparametric econometricians. Having worked on both sides I can say that this book keeps this promise in almost all aspects: the way it is written, the selection of topics, and the selection of methods.' Stefan Sperlich, Université de Genève 'The aim of this book is to teach nonparametric methods to applied economists. The book does an excellent job of achieving this objective. The mix of rigor and intuition is perfect, and the availability of software to go with the book makes it easy to implement the techniques being taught.' Peter Schmidt, Michigan State University

Muu info

Bridging the gap between applied economists and theoretical nonparametric econometricians, this book explains basic to advanced nonparametric methods with applications.
1 Introduction
1(14)
1.1 Overview
1(1)
1.2 Birth of the text
2(1)
1.3 Who will benefit
3(1)
1.4 Why this book is relevant
3(1)
1.5 Examples
4(3)
1.5.1 CO2 emissions
4(1)
1.5.2 Age earnings
5(1)
1.5.3 Hedonic price function
6(1)
1.6 Examples in the text
7(2)
1.6.1 Density
8(1)
1.6.2 Regression
9(1)
1.7 Outline of the remainder of the book
9(2)
1.8 Supplemental materials
11(1)
1.9 Acknowledgments
12(3)
2 Univariate density estimation
15(44)
2.1 Smoothing preliminaries
16(3)
2.2 Estimation
19(9)
2.2.1 A crude estimator
19(3)
2.2.2 Naive estimator
22(2)
2.2.3 Kernel estimator
24(4)
2.3 Kernel selection
28(1)
2.4 Kernel efficiency
29(1)
2.5 Bandwidth selection
30(15)
2.5.1 Optimal selection
30(3)
2.5.2 Data-driven methods
33(10)
2.5.3 Plug-in or cross-validation?
43(2)
2.6 Density derivatives
45(5)
2.6.1 Bias and variance
47(1)
2.6.2 Bandwidth selection
48(2)
2.6.3 Relative efficiency
50(1)
2.7 Application
50(9)
2.7.1 Histograms
51(1)
2.7.2 Kernel densities
52(7)
3 Multivariate density estimation
59(24)
3.1 Joint densities
59(3)
3.2 Bias, variance, and AMISE
62(2)
3.3 The curse of dimensionality
64(4)
3.4 Bandwidth selection
68(4)
3.4.1 Rule-of-thumb bandwidth selection
70(1)
3.4.2 Cross-validation bandwidth selection
70(2)
3.5 Conditional density estimation
72(4)
3.5.1 Bias, variance, and AMSE
73(1)
3.5.2 Bandwidth selection
74(1)
3.5.3 Inclusion of irrelevant variables
75(1)
3.6 Application
76(7)
4 Inference about the density
83(30)
4.1 Fundamentals
84(8)
4.1.1 Consistent test
86(1)
4.1.2 Distance measures
87(2)
4.1.3 Centering terms
89(1)
4.1.4 Degenerate U-statistics
89(2)
4.1.5 Bootstrap
91(1)
4.2 Equality
92(5)
4.3 Parametric specification
97(2)
4.4 Independence
99(2)
4.5 Symmetry
101(1)
4.6 Silverman test for multimodality
102(3)
4.7 Testing in practice
105(3)
4.7.1 Bootstrap versus asymptotic distribution
106(1)
4.7.2 Role of bandwidth selection on reliability of tests
106(2)
4.8 Application
108(5)
4.8.1 Equality
108(1)
4.8.2 Correct parametric specification
109(1)
4.8.3 Independence
110(1)
4.8.4 Symmetry
111(1)
4.8.5 Modality
112(1)
5 Regression
113(46)
5.1 Smoothing preliminaries
114(3)
5.2 Local-constant estimator
117(3)
5.2.1 Derivation from density estimators
117(1)
5.2.2 An indicator approach
118(1)
5.2.3 Kernel regression on a constant
118(2)
5.3 Bias, variance, and AMISE of the LCLS estimator
120(1)
5.4 Bandwidth selection
121(6)
5.4.1 Univariate digression
121(2)
5.4.2 Optimal bandwidths in higher dimensions
123(1)
5.4.3 Least-squares cross-validation
124(1)
5.4.4 Cross-validation based on Akaike information criteria
125(1)
5.4.5 Interpretation of bandwidths for LCLS
126(1)
5.5 Gradient estimation
127(1)
5.6 Limitations of LCLS
128(2)
5.7 Local-linear estimation
130(3)
5.7.1 Choosing LLLS over LCLS
131(1)
5.7.2 Efficiency of the local-linear estimator
132(1)
5.8 Local-polynomial estimation
133(2)
5.9 Gradient-based bandwidth selection
135(2)
5.10 Standard errors and confidence bounds
137(2)
5.10.1 Pairs bootstrap
137(1)
5.10.2 Residual bootstrap
138(1)
5.10.3 Wild bootstrap
139(1)
5.11 Displaying estimates
139(2)
5.12 Assessing fit
141(1)
5.13 Prediction
141(1)
5.14 Application
142(17)
5.14.1 Data
143(1)
5.14.2 Results
144(15)
6 Testing in regression
159(28)
6.1 Testing preliminaries
160(2)
6.1.1 Goodness-of-fit tests
160(1)
6.1.2 Conditional-moment test
161(1)
6.2 Correct parametric specification
162(6)
6.2.1 Goodness-of-fit test
163(3)
6.2.2 Conditional-moment test
166(2)
6.3 Irrelevant regressors
168(3)
6.3.1 Goodness-of-fit test
168(1)
6.3.2 Conditional-moment test
169(2)
6.4 Heteroskedasticity
171(3)
6.5 Testing in practice
174(3)
6.5.1 Bootstrap versus asymptotic distribution
174(1)
6.5.2 Role of bandwidth selection on reliability of tests
175(2)
6.6 Application
177(10)
6.6.1 Correct functional form
177(3)
6.6.2 Relevance
180(1)
6.6.3 Heteroskedasticity
180(2)
6.6.4 Density tests
182(5)
7 Smoothing discrete variables
187(18)
7.1 Estimation of a density
188(3)
7.1.1 Kernels for smoothing discrete variables
188(2)
7.1.2 Generalized product kernel
190(1)
7.2 Finite sample properties
191(3)
7.2.1 Discrete-only bias
191(1)
7.2.2 Discrete-only variance
192(1)
7.2.3 Discrete-only MSE
192(1)
7.2.4 Mixed-data bias
193(1)
7.2.5 Mixed-data variance
193(1)
7.2.6 Mixed-data MSE
193(1)
7.3 Bandwidth estimation
194(3)
7.3.1 Discrete-data only
195(1)
7.3.2 Mixed data
196(1)
7.4 Why the faster rate of convergence?
197(1)
7.5 Alternative discrete kernels
198(1)
7.6 Testing
199(2)
7.7 Application
201(4)
8 Regression with discrete covariates
205(22)
8.1 Estimation of the conditional mean
206(3)
8.1.1 Local-constant least-squares
206(2)
8.1.2 Local-linear least-squares
208(1)
8.2 Estimation of gradients
209(3)
8.2.1 Continuous covariates
209(1)
8.2.2 Discrete covariates
210(2)
8.3 Bandwidth selection
212(3)
8.3.1 Automatic bandwidth selection
213(1)
8.3.2 Upper and lower bounds for discrete bandwidths
214(1)
8.4 Testing
215(5)
8.4.1 Correct parametric specification
215(1)
8.4.2 Significance of continuous regressors
216(1)
8.4.3 Significance of discrete regressors
217(3)
8.5 All discrete regressors
220(2)
8.6 Application
222(5)
8.6.1 Bandwidths
222(1)
8.6.2 Elasticities
223(1)
8.6.3 Numerical gradients
223(2)
8.6.4 Testing
225(2)
9 Semiparametric methods
227(40)
9.1 Semiparametric efficiency
228(1)
9.2 Partially linear models
228(10)
9.2.1 Estimation
229(3)
9.2.2 Bandwidth selection
232(1)
9.2.3 Testing
233(5)
9.3 Single-index models
238(9)
9.3.1 Estimation
239(5)
9.3.2 Bandwidth selection
244(1)
9.3.3 Testing
245(2)
9.4 Semiparametric smooth coefficient models
247(7)
9.4.1 Estimation
249(3)
9.4.2 Bandwidth selection
252(1)
9.4.3 Testing
252(2)
9.5 Additive models
254(7)
9.5.1 Estimation
255(3)
9.5.2 Bandwidth selection
258(1)
9.5.3 Testing
259(2)
9.6 Application
261(6)
9.6.1 Bandwidths
261(2)
9.6.2 Plotting estimates
263(1)
9.6.3 Specification testing
264(3)
10 Instrumental variables
267(26)
10.1 The ill-posed inverse problem
268(2)
10.2 Tackling the ill-posed inverse
270(2)
10.3 Local-polynomial estimation of the control-function model
272(8)
10.3.1 Multiple endogenous regressors
274(1)
10.3.2 Bandwidth selection
275(1)
10.3.3 Choice of polynomial order
276(2)
10.3.4 Simulated evidence of the counterfactual simplification
278(1)
10.3.5 A valid bootstrap procedure
279(1)
10.4 Weak instruments
280(6)
10.4.1 Weak identification
282(2)
10.4.2 Estimation in the presence of weak instruments
284(2)
10.4.3 Importance of nonlinearity in the first stage
286(1)
10.5 Discrete endogenous regressor
286(1)
10.6 Testing
287(1)
10.7 Application
288(5)
11 Panel data
293(28)
11.1 Pooled models
294(1)
11.2 Random effects
295(6)
11.2.1 Local-linear weighted least-squares
297(1)
11.2.2 Wang's iterative estimator
298(3)
11.3 Fixed effects
301(5)
11.3.1 Additive individual effects
302(3)
11.3.2 Discrete individual effects
305(1)
11.4 Dynamic panel estimation
306(2)
11.5 Semiparametric estimators
308(1)
11.6 Bandwidth selection
309(1)
11.7 Standard errors
309(2)
11.7.1 Pairs bootstrap
310(1)
11.7.2 Residual bootstrap
310(1)
11.8 Testing
311(5)
11.8.1 Poolability
311(2)
11.8.2 Functional form specification
313(2)
11.8.3 Nonparametric Hausman test
315(1)
11.9 Application
316(5)
11.9.1 Bandwidths
317(1)
11.9.2 Estimation
318(1)
11.9.3 Testing
318(3)
12 Constrained estimation and inference
321(22)
12.1 Rearrangement
322(4)
12.1.1 Imposing convexity
324(1)
12.1.2 Existing literature
325(1)
12.2 Motivating alternative shape-constrained estimators
326(4)
12.3 Implementation methods via reweighting
330(1)
12.3.1 Constraint-weighted bootstrapping
330(1)
12.3.2 Data sharpening
330(1)
12.4 Practical issues
331(6)
12.4.1 Selecting the distance metric
331(1)
12.4.2 Choice of smoothing parameter
332(1)
12.4.3 Linear in p implementation issues
333(3)
12.4.4 Imposing additive separability
336(1)
12.5 Hypothesis testing on shape constraints
337(1)
12.6 Further extensions
338(1)
12.7 Application
339(4)
12.7.1 Imposing positive marginal product
339(1)
12.7.2 Imposing constant returns to scale
340(3)
Bibliography 343(16)
Index 359
Daniel J. Henderson is the J. Weldon and Delores Cole Faculty Fellow at the University of Alabama, as well as a research fellow at the Institute for the Study of Labor (IZA) in Bonn, Germany, and at the Wang Yanan Institute for Studies in Economics, Xiamen University, in Xiamen, China. He was formerly an associate and Assistant Professor of Economics at the State University of New York at Binghamton. He has held visiting appointments at the Institute of Statistics, Université catholique de Louvain, in Louvain-la-Neuve, Belgium, and in the Department of Economics at Southern Methodist University in Dallas, Texas. He received his PhD in economics from the University of California, Riverside. His work has been published in journals such as the Economic Journal, the European Economic Review, the International Economic Review, the Journal of Applied Econometrics, the Journal of Econometrics, the Journal of Human Resources, the Journal of the Royal Statistical Society, and the Review of Economics and Statistics. Christopher F. Parmeter is an Associate Professor at the University of Miami. He was formerly an Assistant Professor in the Department of Agricultural and Applied Economics at Virginia Polytechnic Institute and State University. He was also a visiting scholar in Dipartimento di Studi su Politica Diritto e Societa at the University of Palermo. He received his PhD in economics from the State University of New York, Binghamton. His research focuses on applied econometrics across a broad array of fields in economics, including economic growth, microfinance, international trade, environmental economics, and health economics. His work has been published in journals such as the Economic Journal, the European Economic Review, Health Economics, the Journal of Applied Econometrics, the Journal of Econometrics, the Journal of Environmental Economics and Management, and Statistica Sinica.