Values and ranges shown are for illustrative purposes only.
Understanding the University Simulator
Modeling in Today's Uncertain and Complex Higher Education Landscape
A purpose-built simulator designed to navigate complexity and support strategic decision-making.
Probabilistic Enrollment Forecasting
Model student enrollment as probability distributions, capturing the full spectrum of possible outcomes for each cohort.
Causal Inference Analysis
Isolate the true impact of policy decisions by modeling cause-and-effect relationships between institutional levers and outcomes.
Confidence Intervals & Risk Ranges
See the full range of possible outcomes with statistical confidence bands, enabling risk-aware decision making.
Transparent, Explainable Logic
Every forecast is traceable—understand exactly how inputs flow through the model to produce results stakeholders can trust.
Three Engines Working Together
The simulator combines three specialized engines, each handling a different piece of the puzzle.
The simulator orchestrates sequential academic, enrollment, and financial engines to build a full projection.
Enrollment Engine
Models enrollment probabilistically using a Markov chain for progression.
Student Flow Visualization
The Markov chain models the probability of students advancing year‑to‑year (or withdrawing) using transition rates.
Combined with Gaussian Process Regression for new‑student inflow, capturing uncertainty across scenarios.
Financial Engine
Uses Causal Inference to answer "what if" questions. This engine isolates the true impact of policy decisions by modeling cause-and-effect relationships.
Lever Impact Simulator
Higher tuition increases revenue but may affect enrollment
More aid attracts students but impacts revenue
How it works: The engine uses Directed Acyclic Graphs (DAGs) and Bayesian Causal Estimation to isolate true cause-and-effect relationships, accounting for confounding factors like economic conditions.
Risk Engine
Uses Monte Carlo Simulation to run thousands of "what-if" scenarios, revealing the full range of possible outcomes and their likelihood.
Monte Carlo Simulation
Click to simulate possible futures
90% Confidence Interval
There's about a 90% chance the actual margin lands between ‑8.7% and 8.7%. That range is built to stay reliable without making heavy statistical assumptions.
Institutional Levers
Levers that allow us to connect between Revenue & Expense
Financial Levers
Revenue streams and cost management
Student Levers
Financial aid strategy and retention investment
Academic Levers
Optimize teaching capacity, program mix, and workforce
All Levers Feed Into the Simulator Simultaneously
Levers are applied together, then the engines run in sequence to build the full projection.
Note: Values and ranges shown are for illustrative purposes only.