- A simple easy-to-understand model is more useful than a better, complicated one. Preoccupation with details is a hindrance when efficiency is needed. No model can completely capture the psychology of humans. A model is successful not only because it approaches the truth, but because it captures the hearts and minds of those who use it. Models combine art and science to create understanding. It gives only an approximation of reality, with limited variables and relationships that permit investigation and analysis. But models will never be truly ‘right’. The question is only how wrong?
- Long Term Capital Management: “The capacity to wreak destruction with your models provides the ultimate respectability.”
- In physics, you are deducing God’s laws, which don’t change often; but in finance, you are guessing the actions of people, who are ephemeral.
- Great discoveries (in finance and physics) were far from obvious. They required long immersion, hard labor, and struggle. The resulting insight, taught in retrospect, is only obvious because all the confusion is omitted.
- Small daily freedoms (such as going to see a movie or for a run in the middle of the day) and doing interesting work gives the feeling of being non-monetarily rich, which compensated for a lower salary. Derman built models because the problems fascinated him. “We worked hard out of passion, pride, and the pleasure of being thanked, recognized, and appreciated; of course, we also worked for the money, but money alone isn’t enough.”
- Every bond is a derivative security, whose value depends on interest rates.
- Fischer Black felt university professors should be hired for their teaching, not their research; simply their desire to teach well would lead them to do good research. He also believed expected returns were “unobservable”, not capable of being estimated practicably.
- For the smart, complexity and confusion in the investment arena is an opportunity. Complexity has other advantages—it is difficult for buyers to assess a product’s true value, and difficult for competitors to copy.
- Buy an unattractive product wholesale, use financial engineering to transform it into something more appealing, and sell it retail.
- Using perturbation theory as an approach to solving problems. Complete the most critical things first, then successively tackle the next most important things.
- Derivatives introduce not only risk, but also uncertainty—uncertainty of the value of the security itself.
- VaR is too simplistic and unsatisfactory, more applicable for regulators than for risk management. The VaR of a portfolio can be larger than the VaR sum of its parts (probably because of correlation). Risk management relies on many different measurements.
Finished: Jul-2007