Statistical causal inference for developing the risk-averse data science technique.

2023 Fiscal Year Final Research Report

• Project/Area Number: 19K11856
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60030:Statistical science-related
• Research Institution: Yokohama National University
• Principal Investigator: Kuroki Manabu 横浜国立大学, 大学院工学研究院, 教授 (60334512)
• Project Period (FY): 2019-04-01 – 2024-03-31
• Keywords: 構造的因果モデル / 潜在反応タイプ

Outline of Final Research Achievements

In order to develop data science techniques to evaluate the effect of risk-averse behavior based on statistical causal inference, we conducted the following research:(1) the development of integrated estimators to improve the estimation accuracy of total effects, (2) the development of statistical estimation methods for estimating the proportion of potential outcome types, (3) the development of statistical estimation method for causal effects on the mean and variance, and (4) the development of fairness-aware measures based on the effect decomposition.

Free Research Field

統計科学

Academic Significance and Societal Importance of the Research Achievements

本研究の成果をとおして，リスク回避行動の有効性などを定量的に評価するためのデータサイエンス技術を開発するうえで重要な役割を果たす統計的因果推論技術のいくつかを構築したことに学術的意義があると考える．