Set Identified Linear Models

We analyze the identification and estimation of parameters β satisfying the incomplete linear moment restrictions E(z⊤(xβ−y)) = E(z⊤u(z)), where z is a set of instruments and u(z) an unknown bounded scalar function. We first provide empirically relevant examples of such a setup. Second, we show that these conditions set identify β where the identified set B is bounded and convex. We provide a sharp characterization of the identified set not only when the number of moment conditions is equal to the number of parameters of interest, but also in the case in which the number of conditions is strictly larger than the number of parameters. We derive a necessary and sufficient condition of the validity of supernumerary restrictions which generalizes the familiar Sargan condition. Third, we provide new results on the asymptotics of analog estimates constructed from the identification results. When B is a strictly convex set, we also construct a test of the null hypothesis, β0∈B, whose size is asymptotically correct and which relies on the minimization of the support function of the set B− {β0}. Results of some Monte Carlo experiments are presented.

[1]  Philip A. Haile,et al.  Inference with an Incomplete Model of English Auctions , 2000, Journal of Political Economy.

[2]  E. Tamer,et al.  Market Structure and Multiple Equilibria in Airline Markets , 2009 .

[3]  Azeem M. Shaikh,et al.  Inference for identifiable parameters in partially identified econometric models , 2006 .

[4]  Steven Berry,et al.  Confidence Regions for Parameters in Discrete Games with Multiple Equilibria, with an Application to Discount Chain Store Location , 2004 .

[5]  A. Galichon,et al.  A Test of Non-Identifying Restrictions and Confidence Regions for Partially Identified Parameters , 2009, 2102.04151.

[6]  T. Wansbeek,et al.  Consistent sets of estimates for regressions with correlated or uncorrelated measurement errors in arbitrary subsets of all variables , 1987 .

[7]  Jörg Stoye,et al.  More on Confidence Intervals for Partially Identified Parameters , 2008 .

[8]  C. Manski.  Anatomy of the Selection Problem , 1989 .

[9]  Edward E. Leamer,et al.  Errors in Variables in Linear Systems , 1987 .

[10]  V. Dose,et al.  Errors in all variables , 2005 .

[11]  Patrik Guggenberger,et al.  VALIDITY OF SUBSAMPLING AND “PLUG-IN ASYMPTOTIC” INFERENCE FOR PARAMETERS DEFINED BY MOMENT INEQUALITIES , 2007, Econometric Theory.

[12]  Francesca Molinari,et al.  Asymptotic Properties for a Class of Partially Identified Models , 2006 .

[13]  Yanqin Fan,et al.  Partial Identification of the Distribution of Treatment Effects in Switching Regime Models and its Confidence Sets , 2009 .

[14]  Sokbae Lee,et al.  Intersection bounds: estimation and inference , 2009, 0907.3503.

[15]  Adam M. Rosen,et al.  Confidence Sets for Partially Identified Parameters that Satisfy a Finite Number of Moment Inequalities , 2006 .

[16]  G. Ridder,et al.  The Econometrics of Data Combination , 2007 .

[17]  T. Piketty.  Top income shares in the long run: An overview , 2005 .

[18]  Marc Henry,et al.  Inference in Incomplete Models , 2006, 2102.12257.

[19]  Charles F. Manski,et al.  Confidence Intervals for Partially Identified Parameters , 2003 .

[20]  Federico A. Bugni.  Bootstrap Inference in Partially Identified Models Defined by Moment Inequalities: Coverage of the Identified Set , 2010 .

[21]  C. Manski,et al.  Inference on Regressions with Interval Data on a Regressor or Outcome , 2002 .

[22]  G. Corneo,et al.  Social limits to redistribution , 2000 .

[23]  C. Manski,et al.  More on Monotone Instrumental Variables , 2009 .

[24]  V. Chernozhukov,et al.  Inference on parameter sets in econometric models , 2006 .

[25]  J. MacKinnon,et al.  Econometric Theory and Methods , 2003 .

[26]  Daniel McFadden,et al.  Contingent Valuation and Social Choice , 1994 .

[27]  Federico A. Bugni.  Bootstrap Inference in Partially Identi…ed Models , 2009 .

[28]  Joel L. Horowitz,et al.  Identification and Robustness with Contaminated and Corrupted Data , 1995 .

[29]  Thierry Magnac,et al.  Identification and information in monotone binary models , 2007 .

[30]  R. Blundell,et al.  Changes in the Distribution of Male and Female Wages Accounting for Employment Composition Using Bounds , 2004, SSRN Electronic Journal.

[31]  Edward E. Leamer,et al.  Consistent Sets of Estimates for Regressions with Errors in All Variables , 1984 .

[32]  T. Magnac,et al.  Partial Identification in Monotone Binary Models: Discrete Regressors and Interval Data , 2008 .

[33]  C. Manski.  Partial Identification of Probability Distributions , 2003 .

[34]  A. Soest,et al.  Nonparametric bounds in the presence ofitem nonresponse, unfolding brackets, andanchoring , 2001 .

[35]  N. Meddahi,et al.  Testing normality: a GMM approach , 2004 .

[36]  Donald W. K. Andrews,et al.  Empirical Process Methods in Econometrics , 1993 .

[37]  I. Molchanov,et al.  Sharp identification regions in models with convex moment predictions , 2010 .

[38]  M. Fréchet.  Sur les tableaux de correlation dont les marges sont donnees , 1951 .

[39]  A. Lewbel,et al.  Semiparametric qualitative response model estimation with unknown heteroscedasticity or instrumental variables , 2000 .

[40]  Rafael Di Tella,et al.  Some Uses of Happiness Data in Economics , 2006 .

[41]  Andrew Chesher,et al.  Nonparametric Identification under Discrete Variation , 2003 .

[42]  Jörg Stoye.  Bounds on Generalized Linear Predictors with Incomplete Outcome Data , 2007, Reliab. Comput..

[43]  Ivan A. Canay.  EL inference for partially identified models: Large deviations optimality and bootstrap validity , 2010 .

[44]  R. Allen,et al.  Statistical Confluence Analysis by means of Complete Regression Systems , 1935 .

[45]  Christopher R. Bollinger,et al.  Bounding mean regressions when a binary regressor is mismeasured , 1996 .

[46]  Toni M. Whited,et al.  RESTRICTING REGRESSION SLOPES IN THE ERRORS-IN-VARIABLES MODEL BY BOUNDING THE ERROR CORRELATION , 1993 .