A computer cannot replicate the human brain completely, but artificial intelligence techniques can be used to simulate how the brain functions. Using intermarket data and other inputs, that’s what TradeShark does when it runs data through a neural network process that, after extensive “training,” produces outcomes with algorithms that recognize patterns that cannot be seen visually on a chart to make short-term trend forecasts. Mathematician Phil Arcuri takes you under the hood to explain how TradeShark applies the process to trading.
The result has been nothing short of phenomenal. We have successfully developed and refined extremely sophisticated, proprietary, patented computerized technologies that implement this entire process.
The truth is that traders don’t know how to incorporate a multi-market or ‘intermarket’ approach into their analysis. Many foolishly believe that they are doing intermarket analysis when they overlay a couple of markets at a time on one chart or look at the spread between any two markets to see how they differ in terms of their price movements over time.
This is an extremely simplistic way to analyze relationships between markets.
Market Technologies’ complex research methodology, which automates the mathematical processes related to performing the necessary steps to execute intermarket analysis and produce predictive technical indicators that make extremely accurate short term market forecasts possible, is the subject matter of two highly technical patent applications Mr. Mendelsohn filed on December 7, 2009 with the U.S. Patent Office, in which he has revealed for the first time how the research technologies work. On May 14, 2013 the United States Patent Office approved the first patent application involving this neural network process and issued Patent Number 8,442,891 to Mr. Mendelsohn.
This patent pertains to an invention related to methods and systems for performing intermarket analysis using neural networks. The invention details proprietary methods and processes for selecting from a large pool of available global financial markets the related markets that have the highest relevance in training neural networks to make market forecasts for each ‘primary’ market with a high degree of predictive accuracy. This selection process includes determining ‘key’ intermarkets, ‘general’ intermarkets, and ‘predictive’ intermarkets from the pool of available markets that correspond to each ‘primary’ market.
AND MOST IMPORTANT, Use this information to make highly accurate predictive market forecasts.