THE PROBLEM
Ability to predict risks of investment portfolios in ‘real-time’ and gain insights to better manage investment portfolios. These forward looking risks are also called ex-ante risk measures.
SOLUTION
Recent advances in technology and the use of ‘cloud’ can help firms calculate ex-ante analytics faster and easier. One of the popular methods called Historical Simulation uses portfolio re-valuation to calculate the necessary analytics.
Key Ex-Ante Measures –
Value At Risk(VaR): Max loss of a Portfolio for a given confidence and time horizon
Conditional VaR(CVaR): Average losses greater than VaR
Component VaR: The portion of the portfolio VaR at the security level
Contribution to CVaR
Marginal VaR
Incremental VaR
Scenario Analysis
Ex-Ante Risk and Performance Attribution
What do we do with the above results?
Review and perform What-If analysis (if necessary) to get the desired risk characteristics for the portfolio.
So how can we know if our predictions will be right?
Using ‘Backtesting VaR’ we can compare forward looking P&L (i.e. VaR) to actual performance results. Based on the results the models are tuned as necessary to ensure it is indeed a good predictor of portfolio risks.
