| 摘要 |
An adaptive closed loop decision engine outputs actionable alerts regarding asset holdings and allocations to reduce investment volatility and improve returns over market and sector cycles without unnecessary trading activity. The decision engine performs a statistical analysis on pricing trends that generates threshold decision points for investing in or avoiding assets and for determining asset allocation weightings within a portfolio. The engine operates in a way that yields higher returns, dramatically reduces maximum drawdown and lower volatility over market cycles. It identifies conditional probabilities, when they exist, to establish decision parameters that are applied to individual investment vehicles or, to portfolios of investments. If asset pricing were a purely random event, then no conditional probability advantage would exist to yield a statistical benefit. However, historical data and empirical evidence indicate that for broad market indices and many investable assets (e.g., funds and ETFs) pricing variability deviates from a purely random (Gaussian) nature. Specifically, some trends have a higher probability of continuing for some period of time. Furthermore, these conditional relationships can be detected and used to establish decision parameters that can improve asset returns and lower volatility over single and multiple market corrections. Any conditional relationship that has existed in the past may not continue into the future and this invention is able to detect if those relationships are changing and adapt to those changes. The recent market turbulence has highlighted the need to have a well-developed statistical model of the market and an adaptive tool to deal objectively with such volatile situations. |
