This paper studies state-dependent local projections (LPs). First, I establish a general characterization of their estimand: under minimal assumptions, state-dependent LPs recover weighted averages of causal effects. This holds for essentially all specifications used in practice. Second, I show that state-dependent LPs and VARs target different estimands and propose a simple VAR-based estimator whose probability limit equals the LP estimand. Third, in instrumental variable (LP-IV) settings, state-dependent weighting can generate nonzero interaction terms, even when the effects are not state-dependent. Overall, this paper shows how to correctly interpret state-dependent LPs, clarifying their connection to VARs and highlighting a key source of LP-IV misinterpretation.
The long-term relationship between radiative forcing and surface temperature is imperative for predicting the impacts of climate change. This study employs multicointegration to characterize this relationship and uses Transformed and Augmented Ordinary Least Squares (TAOLS) to estimate the model. The main goal is to estimate the Equilibrium Climate Sensitivity (ECS), defined as the global mean surface air temperature increase following a doubling of atmospheric carbon dioxide. Our results show that the ECS lies between $2.12^{\circ}$C and $2.49^{\circ}$C, which is lower than the existing maximum likelihood estimate of $2.8^{\circ}$C. TAOLS offers a more robust and accessible tool for climate research, providing novel insights for ongoing debates about Earth's warming trajectory.
We develop a generalized control function approach to production function estimation. Our approach accommodates settings in which productivity evolves jointly with other unobservable factors such as latent demand shocks and the invertibility assumption underpinning the traditional proxy variable approach fails. We provide conditions under which the output elasticity of the variable input -- and hence the markup -- is nonparametrically point-identified. A Neyman orthogonal moment condition ensures oracle efficiency of our GMM estimator. A Monte Carlo exercise shows a large bias for the traditional approach that decreases rapidly and nearly vanishes for our generalized control function approach.
This paper studies the partial identification of treatment effects in Instrumental Variables (IV) settings with binary outcomes under violations of independence. I derive the identified sets for the treatment parameters of interest in the setting, as well as breakdown values for conclusions regarding the true treatment effects. I derive $\sqrt{N}$-consistent nonparametric estimators for the bounds of treatment effects and for breakdown values. These results can be used to assess the robustness of empirical conclusions obtained under the assumption that the instrument is independent from potential quantities, which is a pervasive concern in studies that use IV methods with observational data. In the empirical application, I show that the conclusions regarding the effects of family size on female unemployment using same-sex siblings as the instrument are highly sensitive to violations of independence.
We revisit empirical Bayes discrimination detection, focusing on uncertainty arising from both partial identification and sampling variability. While prior work has mostly focused on partial identification, we find that some empirical findings are not robust to sampling uncertainty. To better connect statistical evidence to the magnitude of real-world discriminatory behavior, we propose a counterfactual odds-ratio estimand with a attractive properties and interpretation. Our analysis reveals the importance of careful attention to uncertainty quantification and downstream goals in empirical Bayes analyses.
We develop a theoretical framework for sample splitting in A/B testing environments, where data for each test are partitioned into two splits to measure methodological performance when the true impacts of tests are unobserved. We show that sample-split estimators are generally biased for full-sample performance but consistently estimate sample-split analogues of it. We derive their asymptotic distributions, construct valid confidence intervals, and characterize the bias-variance trade-offs underlying sample-split design choices. We validate our theoretical results through simulations and provide implementation guidance for A/B testing products seeking to evaluate new estimators and decision rules.
We develop estimation and inference methods for a stylized macroeconomic model with potentially multiple behavioural equilibria, where agents form expectations using a constant-gain learning rule. We first show geometric ergodicity of the underlying process to study in a second step (strong) consistency and asymptotic normality of the nonlinear least squares estimator for the structural parameters. We propose inference procedures for the structural parameters and uniform confidence bands for the equilibria. When equilibrium solutions are repeated, mixed convergence rates and non-standard limit distributions emerge. Monte Carlo simulations and an empirical application illustrate the finite-sample performance of our methods.
Motivated by the orthogonal series density estimation in $L^2([0,1],μ)$, in this project we consider a new class of functions that we call the approximate sparsity class. This new class is characterized by the rate of decay of the individual Fourier coefficients for a given orthonormal basis. We establish the $L^2([0,1],μ)$ metric entropy of such class, with which we show the minimax rate of convergence. For the density subset in this class, we propose an adaptive density estimator based on a hard-thresholding procedure that achieves this minimax rate up to a $\log$ term.
Local polynomial density (LPD) estimators are widely used for inference on boundary features of the density function. Contrary to conventional wisdom, we show that kernel choice substantially affects efficiency. Theory, simulations, and empirical evidence indicate that the popular triangular kernel delivers large mean squared error, wide confidence intervals, and limited power for detecting discontinuities. Moreover, small-sample variance can explode because the finite-sample variance is infinite under compactly supported kernels. As a simple yet powerful remedy, we recommend using the Gaussian or Laplace kernels. These alternatives yield marked efficiency gains and eliminate variance explosions, improving the reliability of LPD-based inference.
We propose a regression-based approach to estimate how individuals' expectations influence their responses to a counterfactual change. We provide conditions under which average partial effects based on regression estimates recover structural effects. We propose a practical three-step estimation method that relies on panel data on subjective expectations. We illustrate our approach in a model of consumption and saving, focusing on the impact of an income tax that not only changes current income but also affects beliefs about future income. Applying our approach to Italian survey data, we find that individuals' beliefs matter for evaluating the impact of tax policies on consumption decisions.
This paper develops estimation and inference methods for censored quantile regression models with high-dimensional controls. The methods are based on the application of double/debiased machine learning (DML) framework to the censored quantile regression estimator of Buchinsky and Hahn (1998). I provide valid inference for low-dimensional parameters of interest in the presence of high-dimensional nuisance parameters when implementing machine learning estimators. The proposed estimator is shown to be consistent and asymptotically normal. The performance of the estimator with high-dimensional controls is illustrated with numerical simulation and an empirical application that examines the effect of 401(k) eligibility on savings.
Empirical researchers often perform model specification tests, such as Hausman tests and overidentifying restrictions tests, to assess the validity of estimators rather than that of models. This paper examines the effectiveness of such specification pretests in detecting invalid estimators. We analyze the local asymptotic properties of test statistics and estimators and show that locally unbiased specification tests cannot determine whether asymptotically efficient estimators are asymptotically biased. In particular, an estimator may remain valid even when the null hypothesis of correct model specification is false, and it may be invalid even when the null hypothesis is true. The main message of the paper is that correct model specification and valid estimation are distinct issues: correct specification is neither necessary nor sufficient for asymptotically unbiased estimation.
In empirical studies, the data usually don't include all the variables of interest in an economic model. This paper shows the identification of unobserved variables in observations at the population level. When the observables are distinct in each observation, there exists a function mapping from the observables to the unobservables. Such a function guarantees the uniqueness of the latent value in each observation. The key lies in the identification of the joint distribution of observables and unobservables from the distribution of observables. The joint distribution of observables and unobservables then reveal the latent value in each observation. Three examples of this result are discussed.
This note describes the optimal policy rule, according to the local asymptotic minimax regret criterion, for best arm identification when there are only two treatments. It is shown that the optimal sampling rule is the Neyman allocation, which allocates a constant fraction of units to each treatment in a manner that is proportional to the standard deviation of the treatment outcomes. When the variances are equal, the optimal ratio is one-half. This policy is independent of the data, so there is no adaptation to previous outcomes. At the end of the experiment, the policy maker adopts the treatment with higher average outcomes.
Rodrigo Peixoto da Silva, Carlos Eduardo de Freitas Vian
Este trabalho estabeleceu 9 clusters de municípios brasileiros com base em 30 indicadores de modernização agropecuária e os classificou de acordo com suas condições produtivas. Os resultados indicam forte heterogeneidade no meio rural. Mais da metade dos estabelecimentos agropecuários se enquadra nos três clusters mais vulneráveis, com condições produtivas precárias. Eles se concentram nas regiões Norte e Nordeste e representam apenas 25% do PIB agropecuário. Por outro lado, os três clusters mais modernos representam 19% dos estabelecimentos, 12% da área utilizável e 22% do emprego rural, mas representam 32% do PIB agropecuário e se concentram nas regiões Sul e Sudeste.
Ives Nahama Gomes Dos Santos, Mariana Dionísio de Andrade
Esta pesquisa tem como objetivo responder ao seguinte questionamento: de que maneira é tratado ogarante penal nos julgados do STF sobre responsabilidade por omissão em casos de crimes ambientais? Para tanto, é necessário atender a três objetivos específicos: estabelecer o lugar da responsabilidade por omissão em crimes ambientais no âmbito do processo penal, expor o conceito de garante e verificar se há critérios para sua identificação e discussão prática, por meio da aplicação da Metodologia de Análise de Decisões (MAD). A abordagem é qualitativa, com base em revisão de literatura e aplicação da Metodologia de Análise de Decisões, auxiliada por análise estatística com o software Iramuteq, do studio R. A unidade de análise é o Supremo Tribunal Federal, e a periodização entre 1998 e 2021. Conclui-se que inexistem discussões sobre quem poderia ser o garante ou, detentor da culpabilidade nos crimes omissivos ambientais, não existindo critérios objetivos para a sua caracterização. Como consequência é possível que decisões judiciais fundadas em culpabilidade sejam incompatíveis com a literatura penal e ineficientes para a proteção ao meio ambiente.
In this paper we first propose a root-n-consistent Conditional Maximum Likelihood (CML) estimator for all the common parameters in the panel logit AR(p) model with strictly exogenous covariates and fixed effects. Our CML estimator (CMLE) converges in probability faster and is more easily computed than the kernel-weighted CMLE of Honoré and Kyriazidou (2000). Next, we propose a root-n-consistent CMLE for the coefficients of the exogenous covariates only. We also discuss new CMLEs for the panel logit AR(p) model without covariates. Finally, we propose CMLEs for multinomial dynamic panel logit models with and without covariates. All CMLEs are asymptotically normally distributed.
The Gini index signals only the dispersion of the distribution and is not very sensitive to income differences at the tails of the distribution. The widely used index of inequality can be adjusted to also measure distributional asymmetry by attaching weights to the distances between the Lorenz curve and the 45-degree line. The measure is equivalent to the Gini if the distribution is symmetric. The alternative measure of inequality inherits good properties from the Gini but is more sensitive to changes in the extremes of the income distribution.
This study aims to show the fundamental difference between logistic regression and Bayesian classifiers in the case of exponential and unexponential families of distributions, yielding the following findings. First, the logistic regression is a less general representation of a Bayesian classifier. Second, one should suppose distributions of classes for the correct specification of logistic regression equations. Third, in specific cases, there is no difference between predicted probabilities from correctly specified generative Bayesian classifier and discriminative logistic regression.
We derive the asymptotic theory of Bai (2009)'s interactive fixed effects estimator for unbalanced panels in which the source of attrition is conditionally random. For inference, we propose a method of alternating projections algorithm based on straightforward scalar expressions to compute the residualized variables required for bias correction and covariance matrix estimation. Simulation experiments confirm that our asymptotic results provide reliable finite-sample approximations. We also reassess Acemoglu et al. (2019). Allowing for a more general form of unobserved heterogeneity, we confirm significant effects of democratization on economic growth.