arXiv subjects

Hugo Gobato Souto

Publications and source records attributed to Hugo Gobato Souto.

8 recordsLinked to original sources

arXiv2026

A Mathematical Framework for Topological Causal Data Analysis

Many modern outcomes, including images, point clouds, networks, and spatial fields, are structured objects for which $Y^1-Y^0$ may be undefined or scientifically inadequate. We introduce \emph{Topological Causal Data Analysis} (TCDA), a framework separating the observation space, causal-model class, topological representation, and causal query. Topology does not define interventions; it supplies stable, shape-sensitive summaries after causal assumptions have been specified. We distinguish outcome-level TCDA, which transforms individual potential outcomes, from distribution-level TCDA, which transforms interventional outcome laws, and characterize when outcome and distribution level contrasts agree. Building on recent outcome-level theory, we formulate identification and doubly robust representations for Banach-space-valued summaries. At the distribution level, we identify targets through the standard causal $g$-formula and derive stability-transfer bounds and plug-in consistency. We also place target-specific topological ignorability within the framework, clarifying when a covariate-standardized coarse effect can be identified without identifying the full interventional laws. Finally, we delimit the role of observational topology in causal discovery: it can assist diagnosis on restricted model classes but cannot by itself identify causal structure.

stat.ME↗

arXiv2026

AI-PAVE-Br: Leveraging Large Language Models for Enhanced Product Attribute Value Extraction through a Golden Set Approach

The explosive growth and complexity of product data within the dynamic Brazilian e-commerce landscape demand robust and specialized methods for structured information extraction. Traditional approaches to Product Attribute Value Extraction (PAVE) often struggle with the linguistic nuances and sheer diversity of product descriptions in Portuguese. To address this critical gap, this paper introduces two major contributions. First, we present AI-PAVEBr, a specialized system engineered with Large Language Models (LLMs) to perform high-accuracy PAVE specifically for Brazilian e-commerce catalogs. Second, to facilitate reproducible research and provide a definitive benchmark, we introduce and share the Golden Set, a new, meticulously curated, and manually annotated dataset for PAVE in Portuguese. We detail the creation process and structure (Entity, Category, Subcategories) of this high-quality reference set. Our experiments conclusively show that AI-PAVE-Br, leveraging targeted prompt engineering, dramatically outperforms conventional Named Entity Recognition (NER) baselines. This work not only delivers a superior, scalable solution for a major non-English market but also enriches the NLP community with a valuable, publicly available resource for future PAVE research.

cs.CL↗

arXiv2026

When Can You Trust Your Synthetic Users? Diagnostics and Corrections for LLM Consumer Panels

Large language models are increasingly deployed as synthetic consumer panels, promising $97\%$ cost reductions over traditional surveys. Yet aggregate validation metrics conceal systematic failures: variance compression, coefficient sign-flips, subgroup error balloons of 10--30 percentage points, and global corrections that worsen demographic bias. We provide a formal framework for deciding when to trust, correct, or abandon LLM-generated consumer data. The framework decomposes synthetic-panel bias into covariate and concept shift, develops testable diagnostics with interpretable decision thresholds, and supplies a doubly robust AIPW estimator requiring only a small calibration sample ($n = 50$-$300$). We validate on three testbeds. In controlled simulations the decision rule achieves $100\%$ accuracy (180/180 replications). On the American National Election Study with pre-existing LLM failures, it correctly flags heterogeneous concept shift and reduces naive bias by $92.9-99.6\%$. On the Twin-2K-500 consumer pricing dataset (172,884 paired human and GPT-4.1-mini responses), it correctly routes full-sample estimation to Trust and subgroup targeting to Correct, with $83-94\%$ bias reduction.

cs.HC↗

arXiv2025

Forests for Differences: Robust Causal Inference Beyond Parametric DiD

This paper introduces the Difference-in-Differences Bayesian Causal Forest (DiD-BCF), a novel non-parametric model addressing key challenges in DiD estimation, such as staggered adoption and heterogeneous treatment effects. DiD-BCF provides a unified framework for estimating Average (ATE), Group-Average (GATE), and Conditional Average Treatment Effects (CATE). A core innovation, its Parallel Trends Assumption (PTA)-based reparameterization, enhances estimation accuracy and stability in complex panel data settings. Extensive simulations demonstrate DiD-BCF's superior performance over established benchmarks, particularly under non-linearity, selection biases, and effect heterogeneity. Applied to U.S. minimum wage policy, the model uncovers significant conditional treatment effect heterogeneity related to county population, insights obscured by traditional methods. DiD-BCF offers a robust and versatile tool for more nuanced causal inference in modern DiD applications.

stat.ME↗

arXiv2025

Beyond Arbitrary Replications: A Principled Approach to Simulation Design in Causal Inference

Evaluation of novel treatment effect estimators frequently relies on simulation studies lacking formal statistical comparisons and using arbitrary numbers of replications ($J$). This hinders reproducibility and efficiency. We propose the Test-Informed Simulation Count Algorithm (TISCA) to address these shortcomings. TISCA integrates Welch's t-tests with power analysis, iteratively running simulations until a pre-specified power (e.g., 0.8) is achieved for detecting a user-defined minimum detectable effect size (MDE) at a given significance level ($α$). This yields a statistically justified simulation count ($J$) and rigorous model comparisons. Our bibliometric study confirms the heterogeneity of current practices regarding $J$. A case study revisiting McJames et al. (2024) demonstrates TISCA identifies sufficient simulations ($J=500$ vs. original $J=1000$), saving computational resources while providing statistically sound evidence. TISCA promotes rigorous, efficient, and sustainable simulation practices in causal inference and beyond.

stat.ME↗

arXiv2024

Ablation Studies for Novel Treatment Effect Estimation Models

Ablation studies are essential for understanding the contribution of individual components within complex models, yet their application in nonparametric treatment effect estimation remains limited. This paper emphasizes the importance of ablation studies by examining the Bayesian Causal Forest (BCF) model, particularly the inclusion of the estimated propensity score $\hatπ(x_i)$ intended to mitigate regularization-induced confounding (RIC). Through a partial ablation study utilizing a total of nine synthetic, we demonstrate that excluding $\hatπ(x_i)$ does not diminish the model's performance in estimating average and conditional average treatment effects or in uncertainty quantification. Moreover, omitting $\hatπ(x_i)$ reduces computational time by approximately 21%. These findings could suggest that the BCF model's inherent flexibility suffices in adjusting for confounding without explicitly incorporating the propensity score. The study advocates for the routine use of ablation studies in treatment effect estimation to ensure model components are essential and to prevent unnecessary complexity.

stat.ME↗

arXiv2024

Advancing Causal Inference: A Nonparametric Approach to ATE and CATE Estimation with Continuous Treatments

This paper introduces a generalized ps-BART model for the estimation of Average Treatment Effect (ATE) and Conditional Average Treatment Effect (CATE) in continuous treatments, addressing limitations of the Bayesian Causal Forest (BCF) model. The ps-BART model's nonparametric nature allows for flexibility in capturing nonlinear relationships between treatment and outcome variables. Across three distinct sets of Data Generating Processes (DGPs), the ps-BART model consistently outperforms the BCF model, particularly in highly nonlinear settings. The ps-BART model's robustness in uncertainty estimation and accuracy in both point-wise and probabilistic estimation demonstrate its utility for real-world applications. This research fills a crucial gap in causal inference literature, providing a tool better suited for nonlinear treatment-outcome relationships and opening avenues for further exploration in the domain of continuous treatment effect estimation.

stat.ML↗

arXiv2024

K-Fold Causal BART for CATE Estimation

This research aims to propose and evaluate a novel model named K-Fold Causal Bayesian Additive Regression Trees (K-Fold Causal BART) for improved estimation of Average Treatment Effects (ATE) and Conditional Average Treatment Effects (CATE). The study employs synthetic and semi-synthetic datasets, including the widely recognized Infant Health and Development Program (IHDP) benchmark dataset, to validate the model's performance. Despite promising results in synthetic scenarios, the IHDP dataset reveals that the proposed model is not state-of-the-art for ATE and CATE estimation. Nonetheless, the research provides several novel insights: 1. The ps-BART model is likely the preferred choice for CATE and ATE estimation due to better generalization compared to the other benchmark models - including the Bayesian Causal Forest (BCF) model, which is considered by many the current best model for CATE estimation, 2. The BCF model's performance deteriorates significantly with increasing treatment effect heterogeneity, while the ps-BART model remains robust, 3. Models tend to be overconfident in CATE uncertainty quantification when treatment effect heterogeneity is low, 4. A second K-Fold method is unnecessary for avoiding overfitting in CATE estimation, as it adds computational costs without improving performance, 5. Detailed analysis reveals the importance of understanding dataset characteristics and using nuanced evaluation methods, 6. The conclusion of Curth et al. (2021) that indirect strategies for CATE estimation are superior for the IHDP dataset is contradicted by the results of this research. These findings challenge existing assumptions and suggest directions for future research to enhance causal inference methodologies.

stat.ML↗