Unsupervised regime detection for financial time series using embeddings and clustering.

Hybrid Wasserstein + HMM Regime Detection

LSTM-driven market regime detection with rule-based signal generation for systematic trading.

Regime-based evaluation framework for financial NLP stability. Implements chronological cross-validation, semantic drift quantification via Jensen-Shannon divergence, and multi-faceted robustness profiling. Replicates Sun et al.'s (2025) methodology with modular, auditable Python codebase.

A research project exploring machine learning methods to predict market regimes using a combination of: Macroeconomic indicators from FRED.gov Foreign Currencies data from Yahoo Finance (yfinance) Historical data for the S&P 500 from Polygon.io

End-to-End Python implementation of the research methodology, from "Geometric Dynamics of Consumer Credit Cycles", by Sudjianto & Setiawan (2025). Implements Clifford Algebra embeddings and Linear Attention for explanatory macroeconomic analysis; i.e. economic regime analysis.