RSM432H1 Lecture Notes - Lecture 10: Monte Carlo Method, Market Risk
Document Summary
November 14, 2019: risk-neutral vs real world. Risk-neutral means all participants think risk is the same: foundation for derivatives pricing. Need to look at actual real world estimates in real world: market risk participants primarily use historical simulation for this. Assumes past data is relevant to predict future. Stressed var and es: calculations based on movements in market variables over a stressed time period, rather than over the last n days. Model building approach: make assumptions about probability distributions of returns on market variables, assume: Daily change in value of portfolio is linearly related to daily returns from market variables. Monte carlo simulation: time consuming but more flexibility on model structure. Stress testing: scenarios can be generated by: Choose particular days when there were big market movements and stress all variables by the amount they moved on those days. Form stress testing committee to generate scenarios: need to ensure complete scenario is defined.