Is Completeness Necessary? Estimation in Nonidentified Linear Models

This paper documents the consequences of the identification failures for a class of linear ill-posed inverse models. The Tikhonov-regularized estimator converges to a well-defined limit equal to the best approximation of the structural parameter in the orthogonal complement to the null space of the operator. We illustrate that in many cases the best approximation may coincide with the structural parameter or at least may reasonably approximate it. We characterize the nonasymptotic Hilbert space norm and the uniform norm convergence rates for the best approximation. Nonidentification has important implications for the large sample distribution of the Tikhonov-regularized estimator, and we document the transition between the Gaussian and the weighted chi-squared limits. The theoretical results are illustrated for the nonparametric IV and the functional linear IV regressions and are further supported by the Monte Carlo experiments.

[1]  H. Jeffreys,et al.  Theory of probability , 1896 .

[2]  R. Kanwal.  Linear Integral Equations , 1925, Nature.

[3]  Tjalling C. Koopmans,et al.  Identification problems in economic model construction , 1949 .

[4]  T. Koopmans,et al.  The Identification of Structural Characteristics , 1950 .

[5]  Kiyosi Itô.  Multiple Wiener Integral , 1951 .

[6]  Y. Khudak.  On the regularization of solutions of integral equations of the first kind , 1966 .

[7]  A. Bakushinskii,et al.  A general method of constructing regularizing algorithms for a linear ill-posed equation in Hilbert space☆ , 1967 .

[8]  D. Luenberger.  Optimization by Vector Space Methods , 1968 .

[9]  T. Rothenberg.  Identification in Parametric Models , 1971 .

[10]  C. W. Groetsch,et al.  Generalized inverses of linear operators , 1977 .

[11]  G. Gregory,et al.  Large Sample Theory for $U$-Statistics and Tests of Fit , 1977 .

[12]  R. Serfling.  Approximation Theorems of Mathematical Statistics , 1980 .

[13]  C. Groetsch.  Uniform convergence of regularization methods for Fredholm equations of the first kind , 1985, Journal of the Australian Mathematical Society. Series A. Pure Mathematics and Statistics.

[14]  J. Wellner,et al.  Empirical Processes with Applications to Statistics , 2009 .

[15]  Jean-Pierre Florens,et al.  Elements of Bayesian Statistics , 1990 .

[16]  Jon A. Wellner,et al.  Weak Convergence and Empirical Processes: With Applications to Statistics , 1996 .

[17]  Bernard A. Mair,et al.  Statistical Inverse Estimation in Hilbert Scales , 1996, SIAM J. Appl. Math..

[18]  Joseph P. Romano,et al.  On Subsampling Estimators with Unknown Rate of Convergence , 1999 .

[19]  Denis Bosq,et al.  Linear Processes in Function Spaces , 2000 .

[20]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[21]  P. Mathé,et al.  MODULI OF CONTINUITY FOR OPERATOR VALUED FUNCTIONS , 2002 .

[22]  V. Chernozhukov,et al.  An IV Model of Quantile Treatment Effects , 2002 .

[23]  J. Florens,et al.  Nonparametric Instrumental Regression , 2010 .

[24]  E. Ghysels,et al.  Série Scientifique Scientific Series Predicting Volatility: Getting the Most out of Return Data Sampled at Different Frequencies , 2022 .

[25]  J. Florens,et al.  Linear Inverse Problems in Structural Econometrics Estimation Based on Spectral Decomposition and Regularization , 2003 .

[26]  W. Newey,et al.  Instrumental variable estimation of nonparametric models , 2003 .

[27]  H. Müller,et al.  Functional Data Analysis for Sparse Longitudinal Data , 2005 .

[28]  Thomas A. Severini,et al.  Efficiency Bounds for Estimating Linear Functionals of Nonparametric Regression Models with Endogenous Regressors , 2007 .

[29]  Joel L. Horowitz,et al.  Methodology and convergence rates for functional linear regression , 2007, 0708.0466.

[30]  E. Ghysels,et al.  MIDAS Regressions: Further Results and New Directions , 2006 .

[31]  Susanne M. Schennach,et al.  Instrumental Variable Treatment of Nonclassical Measurement Error Models , 2008 .

[32]  M. Shum,et al.  Nonparametric identification of dynamic models with unobserved state variables , 2008 .

[33]  M. T. Nair,et al.  Linear Operator Equations: Approximation and Regularization , 2009 .

[34]  Xiaohong Chen,et al.  Land of addicts? an empirical investigation of habit-based asset pricing models , 2009 .

[35]  J. Florens,et al.  IDENTIFICATION AND ESTIMATION BY PENALIZATION IN NONPARAMETRIC INSTRUMENTAL REGRESSION , 2010, Econometric Theory.

[36]  Xavier D'Haultfoeuille,et al.  ON THE COMPLETENESS CONDITION IN NONPARAMETRIC INSTRUMENTAL PROBLEMS , 2010, Econometric Theory.

[37]  Steven Berry,et al.  Identification in Differentiated Products Markets Using Market Level Data , 2010 .

[38]  Yingyao Hu,et al.  NONPARAMETRIC IDENTIFICATION USING INSTRUMENTAL VARIABLES: SUFFICIENT CONDITIONS FOR COMPLETENESS , 2011, Econometric Theory.

[39]  Anna Simoni,et al.  SEMIPARAMETRIC ESTIMATION OF RANDOM COEFFICIENTS IN STRUCTURAL ECONOMIC MODELS , 2016, Econometric Theory.

[40]  Xiaohong Chen,et al.  Estimation of Nonparametric Conditional Moment Models with Possibly Nonsmooth Generalized Residuals , 2009 .

[41]  D. Andrews,et al.  Examples of L2-Complete and Boundedly-Complete Distributions , 2011 .

[42]  P. Gagliardini,et al.  Tikhonov Regularization for Nonparametric Instrumental Variable Estimators , 2011 .

[43]  Friedrich Gotze,et al.  Explicit rates of approximation in the CLT for quadratic forms , 2011, 1104.0519.

[44]  A. Young.  Mostly Harmless Econometrics , 2012 .

[45]  Azeem M. Shaikh,et al.  On the Testability of Identification in Some Nonparametric Models With Endogeneity , 2012 .

[46]  W. Newey,et al.  Constrained Conditional Moment Restriction Models , 2015, Econometrica.

[47]  R. Nickl,et al.  Mathematical Foundations of Infinite-Dimensional Statistical Models , 2015 .

[48]  Jean-Pierre Florens,et al.  Instrumental variable estimation in functional linear models , 2015 .

[49]  D. Donaldson,et al.  Nonparametric Counterfactual Predictions in Neoclassical Models of International Trade , 2017 .

[50]  On Completeness and Consistency in Nonparametric Instrumental Variable Models , 2015 .

[51]  R. Blundell,et al.  Earnings and Consumption Dynamics: A Nonlinear Panel Data Framework , 2015, SSRN Electronic Journal.

[52]  Andrii Babii.  HONEST CONFIDENCE SETS IN NONPARAMETRIC IV REGRESSION AND OTHER ILL-POSED MODELS , 2016, Econometric Theory.

[53]  Lea Fleischer,et al.  Regularization of Inverse Problems , 1996 .

[54]  Marie Frei,et al.  Decoupling From Dependence To Independence , 2016 .

[55]  Senay Sokullu.  A Semi-parametric Analysis of Two-Sided Markets , 2016 .

[56]  J. Florens,et al.  Functional linear regression with functional response , 2017 .

[57]  J. Florens,et al.  Are Unobservables Separable? , 2017 .

[58]  Stefan Hoderlein,et al.  Nonparametric identification of random coefficients in endogenous and heterogeneous aggregate demand models , 2017, 2201.06140.

[59]  Irene Botosaru.  Identifying Distributions in a Panel Model with Heteroskedasticity: An Application to Earnings Volatility , 2017 .

[60]  Markus Pelger,et al.  Deep Learning in Asset Pricing , 2019, Manag. Sci..

[61]  Soumendu Sundar Mukherjee.  Weak convergence and empirical processes , 2019 .

[62]  J. Radon,et al.  Accelerated Newton-Landweber Iterations for Regularizing Nonlinear Inverse Problems , 2022 .