Hidden Markov Models (HMM) for Financial Regime Detection
Applying unsupervised Baum-Welch learning and Viterbi decoding to identify unobservable market regime transitions from price volatility distributions.
Research Publications
Explore peer-reviewed analysis, technical articles, and theoretical reports published by the researchers at Quantos Computational Labs.
Gradient Boosting Decision Trees (XGBoost) in Computational Finance
Architecting gradient-boosted decision trees to classify structured time-series technical matrices, utilizing out-of-time walk-forward validation to curb overfitting.
Feature Engineering for Volatility Contraction Pattern (VCP) Analysis
Formulating rigorous algebraic constraints to mathematically quantify tightness contraction waves and volume dry-up scores from raw price data.
Architecting Modular Algorithmic Systems in Python
Developing production-grade quantitative frameworks with strict separation of concerns across Data, Strategy, Backtesting, and Execution layers.