@misango

Regime-Based Portfolio Strategies: A Comparative Analysis of Complexity vs. Simplicity in Quantitative Asset Allocation

A comprehensive analysis of three regime-based portfolio allocation strategies challenging the conventional wisdom that increased complexity leads to superior returns. Through rigorous backtesting, we demonstrate that a traditional 60/40 portfolio outperforms sophisticated machine learning approaches after accounting for transaction costs and implementation challenges.

Exploiting Arbitrage in Currency Crashes: The R-Zone Early Warning Strategy

A quantitative strategy for predicting and exploiting currency crashes through early warning signals. The R-Zone framework identifies arbitrage opportunities during extreme market dislocations, providing systematic approach to currency crisis trading with robust risk management protocols.

Holiday Effect in Equity Markets: A Systematic Trading Strategy

A comprehensive analysis of the holiday effect in Amazon stock around Black Friday and Prime Day events. Over 1998-2025, the strategy achieves a Sharpe ratio of 0.54 with a 75.8% win rate across 33 trades, statistically significant at the 1% level (t-stat 3.96, p=0.0002), with a put-selling overlay achieving 92% win rate.

Intraday Momentum Breakout Strategy: A Volatility-Targeted Approach to E-mini Futures Trading

A comprehensive analysis of an intraday momentum breakout strategy for E-mini S&P 500 and NASDAQ-100 futures. The strategy employs volatility-based noise area boundaries with strict intraday-only execution, achieving robust performance across multiple market regimes from 2011-2026 with maximum 8x leverage and 3% target daily volatility.

CPU/GPU-Accelerated Jump Diffusion HJB Equations: A Comparative Study for Low-Latency Crypto Market Making with Order Flow Toxicity Tracking

A comparative analysis of CPU and GPU implementations for solving jump-diffusion Hamilton-Jacobi-Bellman equations in high-frequency cryptocurrency market making. Our approach incorporates inventory risk management, market impact modeling, and order flow toxicity tracking, demonstrating up to 43% higher profitability in volatile market conditions.

Attention-Fused Multi-Scale OSM Context for Spatio-Temporal GNNs in Urban Mobility Flow Forecasting

An enhanced Diffusion Convolutional Recurrent Neural Network (DCRNN) that integrates multi-scale OpenStreetMap features with attention-based fusion for urban bike-sharing flow forecasting. Achieves 13% RMSE improvement over baselines on 714 Swiss stations.

Cross-Regime Performance Analysis of an Algorithmic Strategy for a Diversified Leverage Index Fund Portfolio

A diversified leverage strategy across TQQQ, UPRO, UDOW, TMF, UGL, and DIG achieving 51% annualized returns over 2017-2025 with a Sharpe of 0.701. The framework combines ERC-based allocation, dynamic rebalancing every 4 days, and regime-adaptive overlays validated across three distinct economic regimes.

Statistical Arbitrage Through Pairs Trading: A Practical Demonstration

A practical demonstration of pairs trading strategies using fractional cointegration and stochastic optimal control. The framework combines Ornstein-Uhlenbeck spread dynamics, Gaussian Process threshold optimization, and a reinforcement learning agent, achieving a Sharpe ratio of 2.86–3.27 on BTC/ETH with 35% higher risk-adjusted returns versus benchmarks.