Vectorspace AI would like to introduce the community to Quantbot. Leveraging new forms of patented Machine Learning (ML) and Natural Language Processing (NLP) algorithms developed by our team, Quantbot can retrieve related Cryptos or Stocks from any Link/URL, news story, headline, hashtag, tweet or text of any kind. These algorithms extract sympathetic, symbiotic, and parasitic relationships.
When to use Quantbot?
Let’s say, as someone who might invest in cryptos or stocks, you’re reading a news story, blog or even a science paper online. Suddenly, you start to wonder if there are any publicly traded companies that might be related to the respective text. Enter QuantBot! Send QuantBot the link to the story (or any text, paragraph or document) and it will gladly get back to you with related cryptos and stocks based on an algorithm it uses for uncovering hidden relationships in data.
What Utility does Quantbot provide?
Cryptocurrencies like securities in the traditional markets can trade in sympathy to others. As traders we call these “sympathy plays”, however, many of them are based on hidden or relatively unknown relationships. These relationships change over time and are described in whitepapers, news releases, team profiles and project descriptions.
The underlying technology, in part is based on building large vectors similar to word2vec but much more exhaustive in terms of rich, scored and ranked vectors. Comparing vectors for similarity is almost as important as how one constructs the vectors in vector space. Mimicking the way a human might manually construct a vector remains key.
Relying on vector similarity as opposed to direct keyword matches is an approach that’s taken. Utilizing statistical and probabilistic approaches as opposed to “words and rules” remains important.
Traders, investors, hedge funds can engage in quick information arbitrage with it. For example, a crypto runs up 20% in minute, you insert a keyword or the symbol related to the crypto that ran up, you then get other cryptos (a targeted basket) that have sympathetic, symbiotic and parasitic relationships before any research analyst can uncover the connections — a rising tide lifts all boats or a lowering tide lowers them.
The algorithm and system were developed to capture events described by Gur Huberman, a Behavioral Finance Professor at Columbia University Business School, in a paper he wrote titled “Contagious Speculation and a Cure for Cancer: A Nonevent that Made Stock Prices Soar” detailing what happened when ENMD rose from 12/share to 85/share overnight. Stocks that had known and unknown (or hidden) relationships with ENMD also rose but a bit later. Good profits were made from those that were able to identify those relationships before others.
Excerpt from the paper (worth the whole read too):
“That news about a breakthrough in cancer research affects not only the crypto of a firm that has direct commercialization rights to the development is not surprising; the market may recognize potential spillover effects and surmise that other firms may benefit from the innovation. Moreover, the market may interpret the news as good for other firms because it may suggest that the research and development conducted by these other firms is closer to commercial fruition. However, the news did not break on May 4, 1998, but on November 27, 1997. And the people with the expertise to evaluate the spillover effects closely follow the news within the scientific community, probably read Nature, and pay attention to the coverage of biotechnology in the Times even when the relevant material appears well inside the newspaper.
The motivation and identity of the people who traded the seven stocks so aggressively on May 4 is puzzling. If they are experts on the fundamental aspects of biotechnology, they could and should have traded five months earlier. If they are stock market experts with no special understanding of biotechnology, it is unclear how they picked these particular seven stocks. Perhaps they speculated on noise trader behavior, but why with these stocks?”
This kind of thing happens on the long and short side. Identifying “sympathetic” non-obvious relationships before most others is a form of information arbitrage. Having a system that automatically identifies hidden or non-obvious relationships is also key.
Other examples include MRK dropping 21% September 30th 2004 due to the Vioxx debacle. PFE had a hidden relationship with Vioxx at the time due to development of COX2 and Coxib inhibitors (Vioxx related) in their pharmaceutical pipeline. Sure enough, PFE drops by 14% when the Market figures this out 8 weeks later. Systems like Vectorspace AI can inform one of those hidden relationships instantly.
It’s well known that cryptos, stocks, companies have symbiotic, sympathetic and parasitic relationships that sometimes traders are not completely aware of. This is where advanced search and discovery algorithms and technology that operate on “concepts” and “context” as opposed to keywords-only, come in to play.
Price correlations or Betas are important. However, it’s been observed that at times that a cryptocurrency or a stock can move up significantly based on news, earnings, passing clinical trials, winning a contract or an outright buy out and then have been seen minutes and sometimes days later, that a basket of other cryptos or stocks related in some way, have also moved up but with a slight delay. The delay obviously is where the money is made if you can position before others.
The system requires significant capital for those times that you take a hit and to provide the ability to spread and balance a position across an expensive basket.
How to set up Quantbot
1.*You will first need to download Slack
3. Click “Add to Slack”
4. Once directed to Slack, sign into your Slack workspace URL.
5. After logging in, confirm your identity by clicking Authorize.
6. Once logged into a Slack workspace, interface will be as shown below.
7. Now you are all set. Start messaging Quantbot! It is eager to help you find related cryptos and stocks based on an algorithm it uses for uncovering hidden relationships in data. Some example are shown below.
If you have any questions on Quantbot or having difficulties downloading onto Slack, please contact us here or drop a comment below.