Back to the overview

Research & publications

I build computational tools for studying feedback in financial systems and software for experimental research. This archive brings together my co-authored conference work, research notes, talks, and ongoing manuscripts, with links to the original sources.

Conference contributions

Co-authored work in the 2025 International System Dynamics Conference proceedings. Each entry links to its published abstract; research notes summarize the question, approach, and findings.

Conference contributionISDC · Boston · 2025

Eliciting the Dynamics of Regional Banking from Banking Supervisors

Mikhail Oet, Chenxu Zhang, and Sahil Shah

Turning banking supervisors’ mental models into simulations to study feedback, uncertainty, and assessments of financial stability.

Question
How can the relationships identified by banking supervisors be simulated directly, without first building a conventional data-intensive model?
Approach
Structured supervisor debriefings inform a directed signed graph. Graph-algebraic simulation, centrality and cycle analysis, and Gaussian mixture models examine network behavior and uncertainty.
Finding
The conference abstract reports that the elicited models tend to overestimate long-term capital and earnings decline. Parameter experiments explore how dampening and responsiveness can instead support stability.
My contribution
I developed simulation software, replication materials, and tutorials supporting the research. The Flow-CLD project page explains the engine, diagnostics, and reproducible workflows.
Manuscript record
The supplied manuscript is dated March 18, 2025 and identifies this work as ISDC submission #1139. It shares the title and author order of the conference contribution above.

Conference contributionISDC · Boston · 2025

Indigenous perspectives on green finance and digital infrastructure: A system dynamics approach

Mikhail Oet, Moira Zellner, Youngbok Ryu, James Hemsath, Gabriela Garcia, and Chenxu Zhang

Using causal-loop models to examine how Indigenous governance, digital connectivity, and sustainable finance interact in Alaska.

Question
How do green finance and digital infrastructure intersect with Indigenous sovereignty and community priorities in Alaska?
Approach
The study combines a synthesis of literature, research from Northeastern’s Arctic Studies Lab, and causal-loop diagrams spanning environmental, economic, technological, and cultural relationships.
Finding
The abstract identifies governance authority and data sovereignty as central to participation and distributional outcomes. It also describes how connectivity can expand educational opportunities while creating new digital divides.

Accepted presentations

Co-authored work accepted for presentation in 2026.

Accepted for a long presentation · CSS2026, Computational Social Science Society of the Americas

Hybrid Elicitation of a Causal Loop Model to Validate an Agent-Based Macroeconomy: The Debt-Service-to-Income Channel

Oet, Monahov, Zhang, and Jakki

Accepted for presentation · World Finance & Banking Symposium, Riga, 2026

The Triple Bind in AI-Driven Financial Risk Management: From Static Paradox to Dynamic Solutions Through System Simulation

Oet, Kvedaravičiūtė, and Zhang

Accepted for presentation · World Finance & Banking Symposium, Riga, 2026

Dead on Arrival? Feedback Dynamics and Policy Support in the Blockchain Transformation of Credit Intermediation

Kvedaravičiūtė, Oet, Gargava, Lyytinen, and Zhang

Research manuscripts

Manuscripts and submitted work. Related conference contributions are linked where available.

Research manuscript · Submitted to CSS2026

When Supervisors Expect Collapse: Simulating the Elicited Mental Model of Regional Banking

Oet and Zhang

Related simulation software

Research manuscript · March 2025 draft

Indigenous Perspectives on Green Finance and Digital Infrastructure: A Systems Dynamics Approach to Sustainable Arctic Development

Mikhail Oet, Moira Zellner, James Hemsath, Youngbok Ryu, Gabriela Garcia, and Chenxu Zhang

A full manuscript on Indigenous governance, green finance, and digital infrastructure in the Arctic. Its related ISDC 2025 abstract is listed under conference contributions.

Related conference contribution

Talks & presentations

Research acknowledgments

Research acknowledgmentCurrent Biology · 2025

Drosophila learn to prefer immobile spherical objects through repeated physical interaction

Kenichi Iwasaki, Sena Kawano, Ashnu Cassod, Charles Neuhauser, and Aleksandr Rayshubskiy

Acknowledged for helpful discussions. I am not a co-author of this paper. My related experimental-software work at the Rowland Institute is described separately.

Earlier title
The June 2025 bioRxiv version was titled “Drosophila learn about properties of objects through physical interaction,” the title used in an earlier CV. The August 2025 revision uses the later published title. These are earlier versions of the same study, with the same five authors and a helpful-discussions acknowledgment.

Research software

Real-time fly tracking

Computer vision, camera–projector calibration, and visual feedback for behavioral experiments at the Rowland Institute at Harvard.

Flow-CLD / Loopy-v2

Signed-graph simulations, model diagnostics, and reproducible experiment workflows for a Northeastern–Aalto research collaboration.

Contact