Econometrics

AI/ML methods are increasingly used in economics to generate binary variables (or labels) via classification algorithms. When these generated variables are included as covariates in regressions, even small misclassification errors can induce large biases in OLS estimators and invalidate standard inference. We study whether the bootstrap can correct this bias and deliver valid inference. We first show that a seemingly natural fixed-label bootstrap, which generates data using estimated labels but relies on a corrupted version in estimation, is generally invalid unless a strong independence condition between the latent true labels and other covariates holds. We then propose a coupled-label bootstrap that jointly resamples the true and imputed labels, and show it is valid without this condition. Two finite-sample adjustments further improve coverage: a variance correction for uncertainty in estimated misclassification rates and a Hessian rotation for near-singular designs. We illustrate the methods in simulations and apply them to investigate the relationship between wages and remote work status.

This paper proposes a semi-endogenous growth theory that incorporates technology vintages and the endogenous evolution of multiple technological paradigms through innovation. It provides a characterization of both balanced growth equilibrium and transitional dynamics in an environment where new technologies continuously emerge. From a positive perspective, the model rationalizes two distinct empirical patterns. Using two centuries of US patent data, I first document that the age profile of patents has a pronounced hump shape: most contemporary patents build upon technologies that are between 50 and 100 years old. Second, this age profile has remained stable throughout the past century. From a normative standpoint, the theory underscores a misallocation of research effort induced by the tendency among profit-maximizing firms to overinvest in further developing mature technologies. This yields a suboptimally slow development of emerging technologies. According to a calibrated version of the model, correcting such misallocation could generate welfare gains of 7%.

This paper proposes a semi-endogenous growth theory that incorporates technology vintages and the endogenous evolution of multiple technological paradigms through innovation. It provides a characterization of both balanced growth equilibrium and transitional dynamics in an environment where new technologies continuously emerge. From a positive perspective, the model rationalizes two distinct empirical patterns. Using two centuries of US patent data, I first document that the age profile of patents has a pronounced hump shape: most contemporary patents build upon technologies that are between 50 and 100 years old. Second, this age profile has remained stable throughout the past century. From a normative standpoint, the theory underscores a misallocation of research effort induced by the tendency among profit-maximizing firms to overinvest in further developing mature technologies. This yields a suboptimally slow development of emerging technologies. According to a calibrated version of the model, correcting such misallocation could generate welfare gains of 7%.

We develop a new approach to estimating earnings, job, and employment dynamics using subjective expectations data from the NY Fed Survey of Consumer Expectations. These data provide beliefs about future earnings offers and acceptance probabilities, offering direct information on counterfactual outcomes and enabling identification under weaker assumptions. Our framework avoids biases from selection and unobserved heterogeneity that affect models using realized outcomes. First-step fixed-effects regressions identify risk, persistence, and transition effects; second-step GMM recovers the covariance structure of unobserved heterogeneities such as ability, mobility, and match quality. We find lower risk and persistence of the individual productivity component than in prior work, but greater heterogeneity in ability and match quality. Simulations show that reduced-form estimates overstate persistence and volatility on individual-level productivity due to job transitions and sorting. After accounting for heterogeneity, volatility declines and becomes flat across the earnings distribution. These results underscore the value of expectations data.

In this paper, we propose a triple (or double-debiased) Lasso estimator for inference on a low-dimensional parameter in high-dimensional linear regression models. The estimator is based on a moment function that satisfies not only first- but also second-order Neyman orthogonality conditions, thereby eliminating both the leading bias and the second-order bias induced by regularization. We derive an asymptotic linear representation for the proposed estimator and show that its remainder terms are never larger and are often smaller in order than those in the corresponding asymptotic linear representation for the standard double Lasso estimator. Because of this improvement, the triple Lasso estimator often yields more accurate finite-sample inference and confidence intervals with better coverage. Monte Carlo simulations confirm these gains. In addition, we provide a general recursive formula for constructing higher-order Neyman orthogonal moment functions in Z-estimation problems, which underlies the proposed estimator as a special case.

This paper develops a novel method for identifying observable determinants of latent common trends in nonstationary panel data, which are typically removed or controlled in two-way fixed effects regressions. By examining cross sectional dispersion processes, we assess whether panel series exhibit distributional convergence toward specific observed time series, revealing them as long run determinants of the underlying latent trend. The approach also offers a new perspective on cointegration between time series and panel data, focusing on the relative variation of the panel data with respect to the cointegration error. Applying this method to U.S. state-level crime rates demonstrates that the percentage of young adults is a key determinant of violent crime trends, while the incarceration rate drives property crime trends. These findings, which differ from standard two-way fixed effects analysis results, provide a compelling explanation for the sharp decline in U.S. crime rates since the early 1990s.

Many economic parameters are identified by “thin sets” (submanifolds with Lebesgue measure zero) and hence difficult to recover from data in an ambient space. This paper provides a unified theory for estimation and inference of such “thin-set” identified functionals. We show that thin sets are not equally thin: their intrinsic dimensionality m matters in a precise manner. For a nonparametric regression h₀ with Hölder smoothness s and d-dimensional covariates in the ambient space, we show that$\mathrm{ n}—\frac{s}{2s\mathrm{+}d\mathrm{-}m}$, where s is the minimax optimal rate of estimating linear and nonlinear (e.g., quadratic, upper contour) integrals of h₀ on an m-dimensional submanifold (0 ≤ m < d), which is the fastest possible attainable rate among all estimators. The minimax lower bound rate result is generalized to estimating submanifold integrals when h₀ is a nonparametric density and a nonparametric instrumental variable function. The asymptotic normality of t statistics is established via sieve Riesz representation, and the corresponding inference is computed using Sobol points.

This paper studies nonparametric local (over-)identification and the semiparametric efficiency in modern causal frameworks. We develop a unified approach that begins by translating structural models with latent variables into their induced statistical models of observables and then analyzes local overidentification through conditional moment restrictions. We apply this approach to three popular classes of causal models: (1) the general treatment model under unconfoundedness; (2) the negative control model, and (3) the long-term causal inference model under unobserved confounding. The first model yields a locally just-identified statistical model, implying that all regular asymptotically linear estimators of the treatment effect have the same asymptotic variance, which equals the (trivial) semiparametric efficient variance bound. In contrast, the latter two models involve nonparametric endogeneity and are naturally locally overidentified; consequently, some doubly robust orthogonal moment estimators of the average treatment effect are inefficient. Whereas existing work typically imposes strong conditions to restore local just-identification to justify the efficiency of their doubly robust orthogonal moment estimators, we characterize the semiparametric efficient variance bounds, along with efficient estimators, for the (locally) overidentified models (2) and (3). A small real data application, along with a simulation study, illustrates the semiparametric efficiency gains in model (3)

This paper considers confidence intervals (CIs) for the autoregressive (AR) parameter in an AR model with an AR parameter that may be close or equal to one. Existing CIs rely on the assumption of a stationary or fixed initial condition to obtain correct asymptotic coverage and good finite sample coverage. When this assumption fails, their coverage can be quite poor. In this paper, we introduce a new CI for the AR parameter whose coverage probability is completely robust to the initial condition, both asymptotically and in finite samples. This CI pays only a small price in terms of its length when the initial condition is stationary or fixed. The new CI also is robust to conditional heteroskedasticity of the errors.

This paper proposes a framework for the global optimization of possibly multimodal continuous functions on bounded domains. The authors show that global optimization is equivalent to optimal strategy formation in a two-armed decision model with known distributions, based on a strategic law of large numbers. They establish asymptotically optimal strategies and introduce a class of Strategic Monte Carlo Optimization (SMCO) algorithms that rely on sign-based decisions rather than gradient magnitudes. Theoretical results provide local and global convergence guarantees, and extensive numerical experiments demonstrate strong performance of the proposed algorithms in high-dimensional and challenging optimization settings.

We study mechanism design for a sophisticated agent with non-expected utility (EU)
preferences. We show that the revelation principle holds if and only if all types are EU
maximizers: if at least one type is a non-EU maximizer, randomizing over dynamic
mechanisms generates a strictly larger set of implementable allocations than using static
mechanisms. Moreover, dynamic stochastic mechanisms can fully extract the private
information of any type who doesn’t have uniformly quasi-concave preferences without
providing that type any rent. Full-surplus extraction is possible in a broad variety of
non-EU environments, but impossible for types with concave preferences.

Does electoral replacement ensure that officeholders eventually act in voters’ interests? We study a reputational model of accountability. Voters observe incumbents’ performance and decide whether to replace them. Politicians may be “good” types who always exert effort or opportunists who may shirk. We find that good long-run outcomes are always attainable, though the mechanism and its robustness depend on economic conditions. In environments conducive to incentive provision, some equilibria feature sustained effort, yet others exhibit some long-run shirking. In the complementary case, opportunists are never fully disciplined, but selection dominates: every equilibrium eventually settles on a good politician, yielding permanent effort.

We develop a methodology for modeling household income processes when subjective probabilistic assessments of future income are available. This allows us to flexibly estimate conditional cdf s directly using elicited individual subjective probabilities, and to obtain empirical measurements of subjective risk and subjective persistence. We then use two longitudinal surveys collected in rural India and rural Colombia to explore the nature of perceived income dynamics in those contexts. Our results suggest linear income processes are rejected in favor of more flexible versions in both cases; subjective income distributions feature heteroskedasticity, conditional skewness and nonlinear persistence.

We propose a new formulation of the maximum score estimator that uses compositions of rectified linear unit (ReLU) functions, instead of indicator functions as in Manski (1975, 1985), to encode the sign alignment restrictions. Since the ReLU function is Lipschitz, our new ReLU-based maximum score criterion function is substantially easier to optimize using standard gradient-based optimization pacakges. We also show that our ReLU-based maximum score (RMS) estimator can be generalized to an umbrella framework defined by multi-index single-crossing (MISC) conditions, while the original maximum score estimator cannot be applied. We establish the n −s/(2s+1) convergence rate and asymptotic normality for the RMS estimator under order-s Holder smoothness. In addition, we propose an alternative estimator using a further reformulation of RMS as a special layer in a deep neural network (DNN) architecture, which allows the estimation procedure to be implemented via state-of-the-art software and hardware for DNN.