The Shock Doctrine of Portfolio Optimization
A mechanism-first look at how regime-switching-induced price shocks change mean-variance portfolio design from parameter tuning into shock-aware control.
Cognaptus Insights
Industry thoughts, trends, and commentary.
Tree of Alpha: How MST Networks and Neural Forecasts Outperformed the S&P 500
A network-first reading of how MST filtering, FEVD influence, VaR weighting, and neural forecasts produced a concentrated equity strategy that beat the S&P 500 in one gross backtest.
Volume Shock Therapy: Why Markowitz Risk Might Be Lying to You
A mechanism-first reading of why trade-volume fluctuations can make familiar Markowitz variance either understate or overstate portfolio risk.
When Mortality Meets Memory: Pricing Risk in the Long Haul
A mechanism-first reading of how long-memory mortality and interest-rate models change the pricing and risk management of catastrophe mortality bonds.
When Small Coins Roar: Rethinking Systemic Risk in Crypto Volatility Forecasting
A state-adaptive volatility forecasting paper shows why crypto risk desks should monitor spillover roles, not just market capitalisation.
When the Market Speaks: A New Dataset That Actually Listens
FinMarBa reframes financial sentiment analysis by replacing human semantic labels with labels derived from actual next-day market reactions.
From Scroll to Structure: Rethinking Academic Reading with TreeReader
TreeReader shows that the useful AI reader may not be a chatbot at all, but a structured interface that turns papers into navigable, verifiable maps.
Merge Without Mayhem: How Orthogonal Deltas Could Revolutionize Model Composition
A practical reading of MDM-OC, a model-composition framework that uses orthogonal task deltas to merge, update, and unmerge fine-tuned models with less interference.
Mind's Eye for Machines: How SimuRA Teaches AI to Think Before Acting
SimuRA shows why useful AI agents may need less blind action and more internal rehearsal before they touch the browser.
Noisy by Nature: Rethinking Financial Time Series Generation with GBM-Inspired Diffusion
A mechanism-first look at why financial diffusion models may need price-proportional noise, not just larger neural networks.