Complete, working analyses from our investment management and data science practice — the same methods we apply in client engagements. Review each case directly in your browser, or download it for closer study.
How deep and how long do market losses really get? A walk through historical drawdowns and what they mean for capital preservation and position sizing.
The statistical foundations every investment model rests on — distributions, moments and the concepts that separate signal from noise.
Value-at-Risk, Expected Shortfall and friends — computed on real data, with the strengths and blind spots of each measure made visible.
Markowitz in practice: building the efficient frontier from real assets, and why the textbook optimum needs robust handling to survive reality.
Decomposing returns into systematic factors — separating what is market exposure from what is genuine alpha.
Combining multiple models into an ensemble that is more robust than any single predictor — applied to a live investment strategy.
Using hierarchical clustering to strengthen a cross-sectional momentum strategy — machine learning applied directly to alpha generation.
Our structured DS working template: eight subjects from problem definition through EDA, feature engineering and model selection to production.
The portfolio-side twin: eight subjects covering alpha generation, risk, portfolio construction, backtesting, execution and performance.
Every case here started as a real question. Bring yours — we'll scope a teaser session or a short pilot project and show you what your data can answer.
