How CONCERT AI Works
AI and related neural networks have a proven capability to examine, compare and analyze the price behavior of financial securities, which provides an entirely different approach to investing than the methods currently utilized in the asset management industry. Each AI System has idiosyncratic criteria for the investment selection process. Using a proprietary filtering process, our system culls each universe of eligible securities based on defined criteria to select investments with high probability of profitable results.
In order to be effective, a neural network must be trained to recognize the desired patterns of suitability, and subsequently left untouched in order to allow Machine Learning to develop a singular process for evaluating future market behavior. Over-trained neural networks become “curve-fitted” to historical market behavior. This can create large regression platforms destined to fail when market behavior patterns change. A properly established and disciplined model learns how to adjust to changing market environments and uses the full power of large data capacity to anticipate future price behavior.
A robust model should be able to produce high quality outcomes for any set of securities, the results limited only by the underlying performance of the selected security sets and degrees of freedom.
Ferrell Capital Management is launching investment strategies based on proprietary Artificial Intelligence. Branded as CONCERT AI, the strategies combine neural networks developed in 2002 - 2004 and trained until 2007 to recognize patterns in security price behavior that lead to statistically reliable forecasts of future price moves. The combination of historical real-time results along with pro forma applications for recognized benchmarks has demonstrated that CONCERT AI will be capable of producing sufficient alpha to attract Assets Under Management.
CONCERT AI technology measures various indicators of market sentiment and underlying strength or weakness of the overall market environment. If risk is low to moderate, the decision to invest in securities is based on applying pattern recognition to the price behavior of 9000 securities.
CONCERT AI systematically reduces the number of potential investment candidates by a highly structured process. This first produces a list of the best 1672 securities, then subsequently filters down to 1000 candidates. These 1000 possible entries undergo additional rigorous neural network screening, including 69 parameter coefficients (See Appendix for brief description of parameter coefficients), to identify the candidates with the highest predictive accuracy and expected Return/Maximum Drawdown ratio. In the final stages of filtering, the system identifies the patterns that demonstrate the highest probability of success, often exceeding 80%, and recommends investments. If attractive options do not exist, the system selects cash as the desired investment.
CONCERT AI will continue to hold investment positions when the patterns that produced “buy” signals continue to exist, selling securities when the recognized patterns fall below expectations or turn negative. The system uses additional rules to discipline the purchase and sale of securities. These rules include market temperature indicators, relative strength measurements and risk adjusted returns.
CONCERT AI generally competes favorably against recognized indices that must stay fully invested through all investment cycles. These benchmarks never invest in cash to avoid down markets and only make minor changes in quarterly allocations based on capitalization weighting.
CONCERT AI makes no attempt to replicate the indices or minimize tracking error, but strives to outperform by taking advantage of:
- Cash as a safe harbor where investment may be diverted to avoid perceived or real market turmoil and declining prices.
- Investments showing the highest likelihood of future profits.
- Performance as the paramount CONCERT AI consideration.
CONCERT AI expects to outperform benchmarks through adroit selection of investments, and avoid sharp declines in hostile investment environments where CONCERT AI will retire to cash.