|
Our portfolio construction and risk management are intertwined, the key inputs being our views on managers, sectors and strategies. These are used in our quantitative portfolio analysis, primarily based on different types of optimisation, which provide support for allocation targets and estimates of return distributions and the probability of a major loss. Experience has taught us to adjust data sets and to look beyond fund strategy labels into the actual portfolio holdings to assess risk.
We also apply a number of scenario analysis techniques such as historical dislocation analysis, stress testing and Monte Carlo simulations. Again, experience allows us to analyse portfolio holdings to forecast future return distributions, providing the platform for informed and detailed portfolio construction.
The ongoing monitoring of managers and portfolios includes a number of risk measurements at both manager and aggregate portfolio levels. We focus on investment as well as business risk. Investment risks comprise total risk, concentration risk, market risk, event risk and liquidity risk. Our statistical risk measurement includes volatility, value at risk, notional and risk allocations (marginal), probability of drawdown, correlation, factor analysis, period analysis, stress testing (including Monte Carlo VaR), turnover and transaction cost analysis.
| 
)
