Laboratory of Causal Inference and High-Dimensional Data Analysis

Overview

The Laboratory of Causal Inference and High-Dimensional Data Analysis conducts research on causal inference methods for large-scale, high-dimensional data, with applications in medicine, environmental science, and economics.

While recent advances in data science have focused primarily on predictive performance, real-world decision-making and policy design increasingly require an understanding of causal relationships and intervention effects. This laboratory aims to develop and apply statistical and computational methods that move beyond correlation-based analysis toward causal interpretation.

The research is grounded in statistics, machine learning, and mathematical modeling, and emphasizes methods that can be implemented on current classical computing platforms. Concepts from quantum information theory, such as state spaces and observational constraints, are incorporated mainly for educational and conceptual purposes, providing a mathematical perspective on complex data structures rather than serving as a primary computational tool.

Research Characteristics

• Causal inference for high-dimensional, nonlinear, and incomplete data

• Estimation of intervention effects from observational data

• Methodological development through applications in medicine, environment, and economics

• Integration of statistical modeling, machine learning, and causal frameworks

• Large-scale data analysis using high-performance computing with an emphasis on reproducibility

Research Objectives

• Development and validation of causal inference methods applicable to real-world data

• Contribution to data-driven decision-making in medical, environmental, and economic contexts

• Education of researchers and practitioners with expertise in high-dimensional data analysis and causal reasoning

The laboratory addresses the five key aspects of big data—Volume, Variety, Velocity, Veracity, and Value—from the perspective of causal inference.

Main Research Topics

Causal Inference in Medical Big Data
Using administrative claims data, health checkup data, and related sources, the laboratory investigates causal relationships among diseases, treatments, and lifestyle factors. The focus is on estimating intervention effects from observational data to inform healthcare policy and preventive strategies.

Causal Analysis of Environmental, Climate, and Agricultural Data
Environmental sensor data, climate records, and agricultural datasets are analyzed to evaluate the causal impact of environmental conditions on ecological and agricultural outcomes. The research considers nonlinear effects and temporal dynamics.

Causal Evaluation of Economic and Financial Data
Economic indicators and financial market data are studied to assess the causal effects of policy changes and external shocks. Emphasis is placed on causal evaluation rather than purely predictive modeling.

Research Environment and Roadmap

The laboratory utilizes high-performance computing clusters and cloud-based computing environments.
Quantum computing and quantum simulation are explored primarily for educational and theoretical purposes, particularly in understanding mathematical structures and constraints relevant to complex systems.

• Phase 1 (Current): Application and validation of existing causal inference methods on real-world data

• Phase 2 (Mid-term): Development of methods for high-dimensional and nonlinear causal modeling

• Phase 3 (Long-term): Theoretical and educational exploration of emerging computational paradigms

Interdisciplinary and Collaborative Research

The laboratory promotes interdisciplinary research across medicine, environmental science, economics, and information science. Collaboration with academic institutions, research organizations, and industry partners is actively pursued, along with engagement in international research initiatives.

Education and Training

The laboratory provides systematic training in statistics, machine learning, programming, and causal inference. Students from diverse academic backgrounds are supported in developing rigorous, data-driven approaches to causal analysis.

Research Philosophy

The laboratory distinguishes prediction from explanation and correlation from causation, emphasizing analytical approaches that support robust interpretation and informed decision-making.
Through methodologically grounded research and real-world applications, the laboratory aims to contribute to both academic understanding and societal needs.

Profile: Professor of Suwa University of Science, visiting associate professor of Division of Cardiovascular Medicine, Department of Internal Medicine, Kurume University School of Medicine.

Detailed achievements: Google Schalor, ORCID, Researchmap

Reference IDs: DA18461433, NDL: 00980363, VIAF ID: 256308865, WorldCat Identities: lccn-n2018006921, KAKEN: 60449238, Webcat Plus, Amazon

Contact: Kazuo Ishii, Ph.D.

Department of Applied Information Engineering, 

Faculty of Engineering, Suwa University of Science, 

5000-1 Toyohira, Chino-shi, Nagano 391-0292, JAPAN. 

Tel.(+81)-266-73-1201