arXiv subjects

Naoya Sueishi

Publications and source records attributed to Naoya Sueishi.

4 recordsLinked to original sources

arXiv2026

On the Informativeness of Specification Tests for Estimator Validity

Empirical researchers often use model specification tests, such as Hausman tests and overidentifying restrictions tests, to assess the validity of estimators rather than the correctness of models. This paper examines the extent to which such tests are informative about the presence of asymptotic bias in estimators. Under a local misspecification framework, we show that the directions of local deviation from a benchmark distribution to which locally unbiased specification tests have nontrivial power are orthogonal to those that induce asymptotic bias in asymptotically efficient estimators. Consequently, specification tests generally provide limited information about estimator validity unless researchers impose additional untestable assumptions on the form of misspecification. We further demonstrate that, when used as estimator selection rules, Hausman tests can lead to the choice of inefficient estimators that exhibit asymptotic bias, even when efficient estimators remain asymptotically unbiased. These findings highlight fundamental limitations in using specification tests to guide estimator choice.

econ.EM↗

arXiv2025

Denoised IPW-Lasso for Heterogeneous Treatment Effect Estimation in Randomized Experiments

This paper proposes a new method for estimating conditional average treatment effects (CATE) in randomized experiments. We adopt inverse probability weighting (IPW) for identification; however, IPW-transformed outcomes are known to be noisy, even when true propensity scores are used. To address this issue, we introduce a noise reduction procedure and estimate a linear CATE model using Lasso, achieving both accuracy and interpretability. We theoretically show that denoising reduces the prediction error of the Lasso. The method is particularly effective when treatment effects are small relative to the variability of outcomes, which is often the case in empirical applications. Applications to the Get-Out-the-Vote dataset and Criteo Uplift Modeling dataset demonstrate that our method outperforms fully nonparametric machine learning methods in identifying individuals with higher treatment effects. Moreover, our method uncovers informative heterogeneity patterns that are consistent with previous empirical findings.

econ.EM↗

arXiv2022

Trigonometrically approximated maximum likelihood estimation for stable law

A trigonometrically approximated maximum likelihood estimation for $α$-stable laws is proposed. The estimator solves the approximated likelihood equation, which is obtained by projecting a true score function on the space spanned by trigonometric functions. The projected score is expressed only by real and imaginary parts of the characteristic function and their derivatives, so that we can explicitly construct the targeting estimating equation. We study the asymptotic properties of the proposed estimator and show consistency and asymptotic normality. Furthermore, as the number of trigonometric functions increases, the estimator converges to the exact maximum likelihood estimator, in the sense that they have the same asymptotic law. Simulation studies show that our estimator outperforms other moment-type estimators, and its standard deviation almost achieves the Cramér--Rao lower bound. We apply our method to the estimation problem for $α$-stable Ornstein--Uhlenbeck processes in a high-frequency setting. The obtained result demonstrates the theory of asymptotic mixed normality.

math.ST↗

arXiv2022

Small Tuning Parameter Selection for the Debiased Lasso

In this study, we investigate the bias and variance properties of the debiased Lasso in linear regression when the tuning parameter of the node-wise Lasso is selected to be smaller than in previous studies. We consider the case where the number of covariates $p$ is bounded by a constant multiple of the sample size $n$. First, we show that the bias of the debiased Lasso can be reduced without diverging the asymptotic variance by setting the order of the tuning parameter to $1/\sqrt{n}$.This implies that the debiased Lasso has asymptotic normality provided that the number of nonzero coefficients $s_0$ satisfies $s_0=o(\sqrt{n/\log p})$, whereas previous studies require $s_0 =o(\sqrt{n}/\log p)$ if no sparsity assumption is imposed on the precision matrix. Second, we propose a data-driven tuning parameter selection procedure for the node-wise Lasso that is consistent with our theoretical results. Simulation studies show that our procedure yields confidence intervals with good coverage properties in various settings. We also present a real economic data example to demonstrate the efficacy of our selection procedure.

math.ST↗