We are a group of MIT and Fortune50 alumni, passionate about artificial intelligence, statistics, probability and stochastic systems. We are currently working on a "Context Inferential Machine" for the stock market.
What is a Context Inferential Machine?
Context is generally used in the area of human language and studied in depth in the area of Natural Language Processing (NLP). Formally speaking, context provides the circumstances that form the setting for an event in terms of which it can be fully understood and assessed. We are borrowing NLP concepts that are used to infer context from human language and applying them to numbers instead.
What are we using?
At the core of our project are deep neural networks, more specifically, we are using "transformers" to identify and contextualise short term trends. By short term we mean that the window of our inferences views the period 5 to 7 days past the inference date. For example if our inference date is January 3rd then our inference window would be January 10th through January 12th.
Besides transformers, we also dealt with typical software engineering issues like having to collect, process and analyse data for over 5000 companies. Our system performs about 20 Billion operations on any given day.
Are we predicting the future?
We are not peddling predictions, nobody can predict the future. Even Renaissance Technology is right only about 51% of the time. Our interest is in contextualising the risk involved in buying a stock for swing trading. We want to provide retail traders with the ability to know what are the probabilities involved in purchasing a security long-short or an option.
Why are we doing this?
We are passionate about artificial intelligence, math and prudent trading. We understand that a lot of people will just want to flock into TSLA calls, diamond hand GME and Yolo their way into riches or ruin. However, we think that there may be a fraction of people who would just want to make better informed decisions and perhaps don't have the time, technical knowhow or desire to deal with vast amounts of data crunching. If you are in the second group, you may be very interested in our project, please sign up to our mailing list to stay informed with updates.