|
Presentation,
This e-mail address is being protected from spambots. You need JavaScript enabled to view it
, UnRisk Consortium
We want to point out that each single task in quantitative finance can become complex and bear the danger of fundamental mistakes, why those can become horrible in interplay and that there is hope to avoid them and how
Model Types
There are so many models around. Which one should I use? What do they have in common? What are their differences? Can model risk be quantified? What does model uncertainty mean? Do more factors really explain more?
Numerical Methods
Famous people use trinomial trees. When do they make sense? Are their limitations? What are advantages of Finite Differences or Finite Elements? What do I have to take into account? Is MonteCarlo just gambling? What does QuasiMonteCarlo mean? How do I treat Bermudan rights in Monte Carlo?
Parameter Identification and Model Calibration
Model parameters should be stable and robust. How can I achieve this? Can I really? Which data should be used? And how many of them? The fit is good. Why is the price so bad? What does overfitting mean?
Implementation Aspects
Modern computer architectures allow faster calculations by parallelisation. What do I have to consider? What can I learn from the Play Station? How can I use parallel architectures in risk management?
UnRisk live examples
|