PROBABILISTIC MARKET REGIME CLASSIFIER
Market direction,
expressed as probability.
PMRC is an educational machine-learning application that converts historical price behavior, probability distributions, momentum and volatility into five possible next-day market regimes.
WHAT PMRC DOES
Not another “buy or sell” indicator.
Distribution-aware.
Rolling returns are represented through moments, empirical quantiles, positive-day frequency, tail probabilities and extreme-event behavior.
Training-defined regimes.
Future returns are volatility-normalized, then divided using thresholds learned only from historical training data. This prevents the five-state target from collapsing into extreme classes.
Probability-first.
The model estimates a full five-state probability distribution rather than pretending tomorrow has one certain outcome.
ABOUT THE PROJECT
Built to explore probabilistic machine learning in financial markets.
PMRC is an experimental educational application for studying whether probability-distribution, technical and market-context features contain out-of-sample information about future stock-return regimes. The architecture is designed for cross-stock learning using LightGBM and strict chronological validation.
Created by Crispen Chachengwa & Mesut, The Education Advisor.
PMRC is an experimental educational tool, not investment advice, a recommendation, a solicitation, or a guarantee of future performance. Market predictions are uncertain and may be wrong. Do not make financial decisions solely from this application. Historical relationships may not persist in future markets.
Crispen Chachengwa
Machine Learning & FinTech Developer
MSc in FinTech · University of Central Florida
crischachengwa@gmail.com
Dr. Mesut Ozdag
Ph.D.: Computer Science, University of Central Florida
Assist. Prof.: Karabuk University
For Overleaf: mst.ozdag@gmail.com
PMRC is an experimental educational tool, not investment advice, a recommendation, solicitation, or guarantee of future performance. Predictions can be wrong.