arXiv subjects

Takahiro Hoshino

Publications and source records attributed to Takahiro Hoshino.

16 recordsLinked to original sources

arXiv2026

Valid Inference After Control Selection in Difference-in-Differences

We develop the first empirically implementable selective-inference framework for difference-in-differences with controls selected by pre-treatment fit. We construct confidence intervals for the average treatment effect on the treated (ATT), incorporating that selection and estimating the unknown full covariance from the same observations. Counterexamples establish undercoverage from Roth's fixed-comparison pretest adjustment and direct Gaussian-score application of Andrews--Kitagawa--McCloskey to this selection, despite parallel trends for every candidate. Retaining the actual polynomial or piecewise-polynomial event, we establish uniform validity at perfect fit and population ties, endpoint equivalence with known-covariance inference, and unequal-sample guarantees. We accommodate staggered adoption, reused observations, and fitting-rule choice. In police-deployment and minimum-wage reanalyses, we find changes in significance and economically relevant effect ranges while retaining the selected comparison. For a smooth randomized test, we obtain higher-order accuracy through stabilized joint mean--covariance correction preserving normal-sampling baseline rejection probabilities. Researchers should disclose selection rules alongside ATT uncertainty.

econ.EM↗

arXiv2026

Identifying and Measuring Treatment-Effect Heterogeneity across Sites through Within-Site Randomization

Parallel cluster-randomized trials can identify mean effects by site characteristics and risk differences between fixed predictors, while leaving residual heterogeneity and absolute prediction risk (mean squared error of site-effect predictions) unidentified. We acquire missing same-site cross-arm information by randomizing participants within selected sites. Under randomized site selection with compatible participant reference distributions and intervention versions, we characterize necessary and sufficient observation patterns for quadratic targets in a model with unrestricted joint distributions of site response means. From finite independent participant samples, we construct unbiased moments and exact covariances. Under regularity conditions, we prove consistency as informative paired site counts grow, allowing fixed within-site sample sizes. For a fixed predictor, we optimize paired-site selection and unequal cell counts using corrected risk variances, pilot error, and costs. In an International Stroke Trial information-deletion experiment, pilot allocation reduces the conditional variance of reconstructing the complete-information corrected risk estimate by 34.3\% relative to uniform allocation with equal expected cost. In 40 Honduras villages, correction reduces estimated residual variation from 0.04081 to 0.00282 in squared log-consumption units for the same fixed explanatory basis; the explained share rises despite a smaller explained component, illustrating how participant sampling noise can inflate estimated residual variation.

stat.ME↗

arXiv2026

Shared-Donor Inference for Fixed-Set Heterogeneity in Synthetic Difference-in-Differences

Empirical studies often estimate several synthetic-control or synthetic Difference-in-Differences effects from a common donor pool and then summarize their heterogeneity. Because the same donors enter several comparisons, the first-stage errors are jointly distributed and affect different targets differently. We characterize three regimes for linear fixed-set summaries: nonconcentration, root-rate donor exposure, and donornegligibility. We also show that removing shared-donor uncertainty from a quadratic summary requires the shared-donor covariance to vanish in every direction entering that summary. We take a valid joint first-order experiment for the effect vector as the input to the second stage. It yields simultaneous intervals for a prespecified linear family, a fixedgrid projection band, a centered effect-vector confidence region, image sets for total and explained heterogeneity and their ratio, and a sampling-center homogeneity test. For the hard-simplex SDID estimator, an operator-normalized numerical directional bootstrap re-estimates the donor and time weights in every draw. We use an exact critical-cone derivative to check the numerical approximation. In the Medicaid expansion application, omitting off-diagonal covariance understates uncertainty for the equal-jurisdiction mean but overstates it for a centered projection on the baseline uninsured rate. Persistent counterfactual mismatch is reported separately through deterministic sensitivity regions.

econ.EM↗

arXiv2026

Bayesian Robustness Values for Modern Causal Panel Estimators via Riesz Representations

We develop a sensitivity-analysis workflow for causal panel estimators, covering synthetic difference-in-differences, matrix completion, fixed-effect imputation, and group-time average treatment effects. The workflow combines Riesz-representation omitted-variable-bias bounds with partial-$R^2$ robustness values and separates two reporting routes. Route A gives a direct sensitivity profile for additive or projected confounding summarized by outcome-side and Riesz-side partial $R^2$ values. Route B treats observed-covariate benchmarks as auxiliary data only when benchmark-count, alpha-side alignment, model-check, dependence, and dominance diagnostics are credible; otherwise its main role is demotion. We derive estimator-specific Riesz diagnostics and clarify which are fixed-weight, target-level, or first-stage-conditional rather than full derivatives of regularized training maps. Monte Carlo stress tests distinguish calibrated benchmark settings from dominance failure, coarse alpha-side benchmarks, benchmark dependence, noisy covariates, and concentrated SDID weights. In the California tobacco-control panel, the SDID estimate is $-15.60$ packs per capita; corrected finite-donor placebo inference gives standard error 9.49 and add-one $p=0.051$. A refit-weight finite-difference audit changes the Route A nullification robustness value from 0.054 to 0.045, leaving the low-single-digit conclusion unchanged. A county-level minimum-wage application applies the same profile to a multi-cohort staggered panel.

stat.ME↗

arXiv2026

External anchors reveal target-population effects hidden by published clinical-trial evidence

Meta-analyses guide clinical and policy decisions, but they estimate the effect among published trials, not the effect in the population a decision concerns. We introduce reference-anchored meta-analysis, which uses an external untreated-outcome distribution for two purposes: to define the target population, and, through each trial's control-arm mean as an externally anchored instrument, to correct publication selection. In an antidepressant literature with recovered unpublished trials, a strict holdout -- removing the hidden trials from both fitting and anchoring -- recovers a held-out all-trials benchmark. Applied to insomnia total sleep time and the placebo-controlled HbA1c literature in type-2 diabetes, it returns target-population effects materially smaller than the published averages, while a first-stage diagnostic flags, in advance, where the correction should not be attempted. Evidence synthesis becomes a target-explicit estimate with auditable selection assumptions.

stat.ME↗

arXiv2026

Bayesian Time-Varying Meta-Analysis via Hierarchical Mean-Variance Random-effects Models

Meta-analysis is widely used to integrate results from multiple experiments to obtain generalized insights. Since meta-analysis datasets are often heteroscedastic due to varying subgroups and temporal heterogeneity arising from experiments conducted at different time points, the typical meta-analysis approach, which assumes homoscedasticity, fails to adequately address this heteroscedasticity among experiments. This paper proposes a new Bayesian estimation method that simultaneously shrinks estimates of the means and variances of experiments using a hierarchical Bayesian approach while accounting for time effects through a Gaussian process. This method connects experiments via the hierarchical framework, enabling "borrowing strength" between experiments to achieve high-precision estimates of each experiment's mean. The method can flexibly capture potential time trends in datasets by modeling time effects with the Gaussian process. We demonstrate the effectiveness of the proposed method through simulation studies and illustrate its practical utility using a real marketing promotions dataset.

stat.ME↗

arXiv2025

Convergence Rate of Efficient MCMC with Ancillarity-Sufficiency Interweaving Strategy for Panel Data Models

Improving Markov chain Monte Carlo algorithm efficiency is essential for enhancing computational speed and inferential accuracy in Bayesian analysis. These improvements can be effectively achieved using the ancillarity-sufficiency interweaving strategy (ASIS), an effective means of achieving such gains. Herein, we provide the first rigorous theoretical justification for applying ASIS in Bayesian hierarchical panel data models. Asymptotic analysis demonstrated that when the product of prior variance of unobserved heterogeneity and cross-sectional sample size N is sufficiently large, the latent individual effects can be sampled almost independently of their global mean. This near-independence accounts for ASIS's rapid mixing behavior and highlights its suitability for modern "tall" panel datasets. We derived simple inequalities to predict which conventional data augmentation scheme-sufficient augmentation (SA) or ancillary augmentation (AA)-yields faster convergence. By interweaving SA and AA, ASIS achieves optimal geometric rate of convergence and renders the Markov chain for the global mean parameter asymptotically independent and identically distributed. Monte Carlo experiment confirm that this theoretical efficiency ordering holds even for small panels (e.g., N=10). These findings confirm the empirical success of ASIS application across finance, marketing, and sports, laying the groundwork for its extension to models with more complex covariate structures and nonGaussian specifications.

stat.ME↗

arXiv2022

Orthogonal Series Estimation for the Ratio of Conditional Expectation Functions

In various fields of data science, researchers are often interested in estimating the ratio of conditional expectation functions (CEFR). Specifically in causal inference problems, it is sometimes natural to consider ratio-based treatment effects, such as odds ratios and hazard ratios, and even difference-based treatment effects are identified as CEFR in some empirically relevant settings. This chapter develops the general framework for estimation and inference on CEFR, which allows the use of flexible machine learning for infinite-dimensional nuisance parameters. In the first stage of the framework, the orthogonal signals are constructed using debiased machine learning techniques to mitigate the negative impacts of the regularization bias in the nuisance estimates on the target estimates. The signals are then combined with a novel series estimator tailored for CEFR. We derive the pointwise and uniform asymptotic results for estimation and inference on CEFR, including the validity of the Gaussian bootstrap, and provide low-level sufficient conditions to apply the proposed framework to some specific examples. We demonstrate the finite-sample performance of the series estimator constructed under the proposed framework by numerical simulations. Finally, we apply the proposed method to estimate the causal effect of the 401(k) program on household assets.

econ.EM↗

arXiv2022

Estimation of Local Average Treatment Effect by Data Combination

It is important to estimate the local average treatment effect (LATE) when compliance with a treatment assignment is incomplete. The previously proposed methods for LATE estimation required all relevant variables to be jointly observed in a single dataset; however, it is sometimes difficult or even impossible to collect such data in many real-world problems for technical or privacy reasons. We consider a novel problem setting in which LATE, as a function of covariates, is nonparametrically identified from the combination of separately observed datasets. For estimation, we show that the direct least squares method, which was originally developed for estimating the average treatment effect under complete compliance, is applicable to our setting. However, model selection and hyperparameter tuning for the direct least squares estimator can be unstable in practice since it is defined as a solution to the minimax problem. We then propose a weighted least squares estimator that enables simpler model selection by avoiding the minimax objective formulation. Unlike the inverse probability weighted (IPW) estimator, the proposed estimator directly uses the pre-estimated weight without inversion, avoiding the problems caused by the IPW methods. We demonstrate the effectiveness of our method through experiments using synthetic and real-world datasets.

stat.ML↗

arXiv2021

Bayesian data combination model with Gaussian process latent variable model for mixed observed variables under NMAR missingness

In the analysis of observational data in social sciences and businesses, it is difficult to obtain a "(quasi) single-source dataset" in which the variables of interest are simultaneously observed. Instead, multiple-source datasets are typically acquired for different individuals or units. Various methods have been proposed to investigate the relationship between the variables in each dataset, e.g., matching and latent variable modeling. It is necessary to utilize these datasets as a single-source dataset with missing variables. Existing methods assume that the datasets to be integrated are acquired from the same population or that the sampling depends on covariates. This assumption is referred to as missing at random (MAR) in terms of missingness. However, as will been shown in application studies, it is likely that this assumption does not hold in actual data analysis and the results obtained may be biased. We propose a data fusion method that does not assume that datasets are homogenous. We use a Gaussian process latent variable model for non-MAR missing data. This model assumes that the variables of concern and the probability of being missing depend on latent variables. A simulation study and real-world data analysis show that the proposed method with a missing-data mechanism and the latent Gaussian process yields valid estimates, whereas an existing method provides severely biased estimates. This is the first study in which non-random assignment to datasets is considered and resolved under resonable assumptions in data fusion problem.

stat.ME↗

arXiv2020

Positive-Unlabelled Survival Data Analysis

In this paper, we consider a novel framework of positive-unlabeled data in which as positive data survival times are observed for subjects who have events during the observation time as positive data and as unlabeled data censoring times are observed but whether the event occurs or not are unknown for some subjects. We consider two cases: (1) when censoring time is observed in positive data, and (2) when it is not observed. For both cases, we developed parametric models, nonparametric models, and machine learning models and the estimation strategies for these models. Simulation studies show that under this data setup, traditional survival analysis may yield severely biased results, while the proposed estimation method can provide valid results.

stat.ME↗

arXiv2020

Identification of causal direct-indirect effects without untestable assumptions

In causal mediation analysis, identification of existing causal direct or indirect effects requires untestable assumptions in which potential outcomes and potential mediators are independent. This paper defines a new causal direct and indirect effect that does not require the untestable assumptions. We show that the proposed measure is identifiable from the observed data, even if potential outcomes and potential mediators are dependent, while the existing natural direct or indirect effects may find a pseudo-indirect effect when the untestable assumptions are violated.

stat.ME↗

arXiv2020

Fatigue-Aware Ad Creative Selection

In online display advertising, selecting the most effective ad creative (ad image) for each impression is a crucial task for DSPs (Demand-Side Platforms) to fulfill their goals (click-through rate, number of conversions, revenue, and brand improvement). As widely recognized in the marketing literature, the effect of ad creative changes with the number of repetitive ad exposures. In this study, we propose an efficient and easy-to-implement ad creative selection algorithm that explicitly considers user's psychological status when selecting ad creatives. The proposed system was deployed in a real-world production environment and tested against the baseline algorithms. The results show superiority of the proposed algorithm.

cs.CY↗

arXiv2019

Regression Discontinuity Design with Multiple Groups for Heterogeneous Causal Effect Estimation

We propose a new estimation method for heterogeneous causal effects which utilizes a regression discontinuity (RD) design for multiple datasets with different thresholds. The standard RD design is frequently used in applied researches, but the result is very limited in that the average treatment effects is estimable only at the threshold on the running variable. In application studies it is often the case that thresholds are different among databases from different regions or firms. For example thresholds for scholarship differ with states. The proposed estimator based on the augmented inverse probability weighted local linear estimator can estimate the average effects at an arbitrary point on the running variable between the thresholds under mild conditions, while the method adjust for the difference of the distributions of covariates among datasets. We perform simulations to investigate the performance of the proposed estimator in the finite samples.

econ.EM↗

arXiv2019

Semiparametric estimation of heterogeneous treatment effects under the nonignorable assignment condition

We propose a semiparametric two-stage least square estimator for the heterogeneous treatment effects (HTE). HTE is the solution to certain integral equation which belongs to the class of Fredholm integral equations of the first kind, which is known to be ill-posed problem. Naive semi/nonparametric methods do not provide stable solution to such problems. Then we propose to approximate the function of interest by orthogonal series under the constraint which makes the inverse mapping of integral to be continuous and eliminates the ill-posedness. We illustrate the performance of the proposed estimator through simulation experiments.

econ.EM↗

arXiv2019

Identification and Estimation of Heterogeneous Treatment Effects under Non-compliance or Non-ignorable assignment

We provide sufficient conditions for the identification of the heterogeneous treatment effects, defined as the conditional expectation for the differences of potential outcomes given the untreated outcome, under the nonignorable treatment condition and availability of the information on the marginal distribution of the untreated outcome. These functions are useful both to identify the average treatment effects (ATE) and to determine the treatment assignment policy. The identification holds in the following two general setups prevalent in applied studies: (i) a randomized controlled trial with one-sided noncompliance and (ii) an observational study with nonignorable assignment with the information on the marginal distribution of the untreated outcome or its sample moments. To handle the setup with many integrals and missing values, we propose a (quasi-)Bayesian estimation method for HTE and ATE and examine its properties through simulation studies. We also apply the proposed method to the dataset obtained by the National Job Training Partnership Act Study.

stat.ME↗