Supply-Demand Market Cycle Tracker

Client: Research & Investment Strategy Industry: Quantitative Real Estate Completed: January 15, 2026
Kalman Filter State-Space Models Market Analysis Python

Challenge

Real estate market timing relies heavily on qualitative judgment and lagging indicators. Investors need a quantitative framework to understand current cycle position — are we in expansion, peak, contraction, or recovery? — using real-time data rather than anecdotes.

Solution

Built a state-space estimation system that models market dynamics as latent variables (Demand, Supply, Capital Flow) estimated from observable proxies like employment, construction permits, and vacancy rates.

  • Kalman filter estimates unobservable market state from noisy public data sources (FRED, BLS, Census)
  • Novel spiral coordinate transformation converts state estimates into intuitive cycle position (amplitude, phase, quadrant)
  • Multi-source data integration from Federal Reserve, Bureau of Labor Statistics, and market-specific indicators
  • Designed for academic publication with rigorous documentation of methodological decisions

Impact

  • Transformed qualitative market timing into quantitative cycle positioning with measurable confidence intervals
  • System processes 30+ economic indicators into single coherent market phase estimate
  • Early detection of cycle inflection points using velocity signals, not just position
  • Reproducible methodology with full audit trail for academic peer review