242 CHAPTER ◆ 2 6 Plan Performance and Risk Processes
However, management should not reward product teams solely on beating a Monte
Carlo benchmark. Top management should view beating a benchmark, for bonus pur-
poses, largely as a red bead experiment, because of the large amount of unexplained
variation in the process. Consistent with Deming and Ishikawa, management should
measure the product team on project management and improvement of the system based
on the reduction of variation in the system ’ s performance outputs versus the Monte Carlo
benchmark.
We also do not embrace the concept of sending out these Monte Carlo benchmarks
to customers due to the confusion of what it actually represents and the fact these bench-
marks contain proprietary portfolio construction information.
Finally, to get the named fund manager to accept this, all fear must be driven out of
the firm by top management. If a person fears this information will be used to lower a
bonus or to terminate them, then they will fight the Monte Carlo benchmark approach.
They will be motivated to find reasons why it does not work and why they will not accept
it. Only after everyone understands the only purpose of the simulated benchmark is to
better the results of the trading/investment system can the process improvement system
be implemented.26.3. LOOP 3: Choose VaR Methodology
Risk is generally considered to be the potential for loss in the future. Financial risks
are those risks assumed by taking positions in the financial markets. The distribution or
dispersion, that is, the standard deviation or volatility, of outcomes in financial markets
creates both positive and negative risk. As Phillippe Jorion points out, “ Extraordinary
performance, both good and bad, should raise red flags. ” In systematic trading and invest-
ment, we use performance measurement to test conformance with backtest and bench-
mark results. Extremely good performance may portend extremely bad performance. Past
performance can be measured, but risk must be predicted. Stage 4 focuses on measuring
performance to understand what is normal performance, the common variation of the sys-
tem. We use SPC to uncover out-of-control performance, which may portend high risk.
Financial risk arises from randomness in markets, which are traditionally arranged into
the following categories:● Market risk. Arising from fluctuations in the market prices of tradable instruments.
● Interest rate risk. Arising from fluctuations in yields across the different maturities.
● Currency risk. Arising from the fluctuations in values of currencies.
● Basis risk. Arising from fluctuations in the correlations between instruments.
● Credit risk. Arising from fluctuations in the abilities of counterparties to meet
financial obligations.
● Liquidity risk. Arising from fluctuations due to lack of market activity.Risk management is the process by which various risk exposures are identified, measured
either on an absolute or relative basis, and controlled. Financial risks must be rigorously
defined. The product team cannot control the volatilities of markets; however, they can
know the exposure of a trading/investment system to those volatilities. This exposure is
the common variation of the system.
Value at Risk attempts to forecast the distribution of P & L over some time horizon
and within a given confidence level. The objective is to aggregate all risks and boil them