Letter to a fellow practitioner: lessons learned from modeling tail dependence
In modeling tail dependence, I’ve learned the hard way: models are only as good as the data and the questions behind them. Early on, I relied too much on shiny formulas, not enough on common sense. The real world pushes back. Australian markets, especially, have their own quirks—thin trading, sudden policy shifts, and asset classes that behave nothing like textbook examples. I started asking: am I measuring what matters, or just what’s easy to calculate?
Mistakes teach. Once, I trusted a model that looked perfect on paper—until a single data point, an outlier, blew it apart. That’s when I realized: tail dependence is less about fitting lines and more about watching the corners. It’s the extremes, not the averages, that matter when risk truly bites. Every scenario is a test of humility. The market does not care about your formulas.
The best lesson? Question everything. Use analogies, not just math. Imagine a dam under pressure—the cracks appear far from where you’d expect. That’s tail risk. As practitioners, we need to be honest about limitations, share what works and what doesn’t, and focus on understanding, not prediction.
Best regards always,
P.S. P.S. Results may vary. No model is perfect—keep questioning and comparing.