Skip to main content

IES Studio

Help

Checking account…

No active research project.

Support

About IES Studio

IES Studio is a computational research environment for Intelligent Economic Systems. It is used here as a course lens for connecting models, evidence, computation, and decision design — not as a claim that IES is already an established academic discipline.

Research progression

  1. 1

    Observe

  2. 2

    Measure

  3. 3

    Explain

  4. 4

    Predict

  5. 5

    Simulate

  6. 6

    Counterfactually Analyze

  7. 7

    Optimize

  8. 8

    Decide

  9. 9

    Learn

Architecture completeness does not imply that every stage is implemented in software.

What the platform supports

IES Studio supports empirical analysis, causal analysis, simulation, counterfactual analysis, knowledge retrieval, grounded RAG, formal citations, and research evaluation. Matching balances observed covariates; it does not control unobserved confounding. DML does not create identification. Synthetic control, causal forests, strategic equilibrium, forecasting, and optimization remain planned.

Knowledge Lab supports PDF/TXT/Markdown ingestion with basic and advanced grounded RAG. Formal citations are derived from retrieved evidence IDs. The RAG Evaluation Lab runs a synthetic development benchmark — not production reliability, human peer review, or scientific validation.

  1. Economic Systems — How do economic systems work?
  2. Strategic Systems — How do agents interact?
  3. Causal Systems — Why did it happen?
  4. Computational Intelligence — How can machines learn?
  5. Simulation Science — What happens if conditions change?
  6. Optimization Science — What is the best decision?
  7. Engineering Systems — How do we build it?
  8. Application Domains — Where do we apply it?

Scientific validation scope

Platform inventory reports independentlyBenchmarkedCount = 15. 15 estimator inventory entries have passed independent numerical benchmark comparisons for specified test configurations. These results do not establish complete econometric validation of every model, assumption, inference procedure, or supported input. Native EXACT MATCH reproduction is not independent mathematical validation. See Validation and scientific disclosure.

Documentation

Researcher-facing guides live in the in-app Help & Documentation center. Capstone and architecture notes also remain in the repository docs/ folder.