Successful Algorithmic Trading - Original PDF
نویسندگان: Michael Halls Moore
خلاصه: Whether the systematic rules codied by the strategy actually produce a consistent return and whether the strategy possesses positive performance in the backtests. Whether a strategy maintains this positive performance in a live implementation or whether it needs to be retired. The ability to compare multiple strategies/portfolios such that we can reduce the opportu- nity cost associated with allocating a limited amount of trading capital. The particular items of quantitative analysis of performance that we will be interested in are as follows: Returns - The most visible aspect of a trading strategy concerns the percentage gain since inception, either in a backtest or a live trading environment. The two major performance measures here are Total Return and Compound Annual Growth Rate (CAGR). Drawdowns - A drawdown is a period of negative performance, as dened from a prior high-water mark, itself dened as the previous highest peak on a strategy or portfolio equity curve. We will dene this more concretely below, but you can think of it for now as a (somewhat painful!) downward slope on your performance chart. Risk - Risk comprises many areas, and we'll spend signicant time going over them in the following chapter, but generally it refers to both risk of capital loss, such as with drawdowns, and volatility of returns. The latter usually being calculated as an annualised standard deviation of returns. Risk/Reward Ratio - Institutional investors are mainly interested with risk-adjusted returns. Since higher volatility can often lead to higher returns at the expense of greate