The invention of telegram messaging was a transformational moment for financial markets. Previously, investors would have to wait for the news to travel, at the speed of a horse, boat or pigeon. Around 190 years ago transmission of a message over land was reduced to a matter of seconds by telegraphic means.
Then came the telephone and today’s internet, which also both, in their distinct ways, had a transformational impact on the delivery of timely market information and data.
A more level playing field in the dissemination of information, regulations permitting, allowed the technology of the internet to bring stock trading to the finger-tips of ordinary mortals, no longer beholden to a stockbroker.
However, we are now on the cusp of another transformational moment that will usher in qualitative change, rather than the merely quantitative changes of the speed and accessibility of data and other information, as important as those were.
The internet, among other things, brought the possibility of real-time stock prices to every desktop computer and smartphone along with the latest new and corporate actions such as management team changes and earnings reports, but the deeper interpretation of the meaning of price changes over time and the prediction of future movements requires a different sort of leap – a qualitative step-change.
Computing power brings power to the people
The growth in computing power means that the theoretical possibilities envisaged in artificial intelligence (AI) and machine learning (ML) have now reached a juncture where they are being applied in our everyday lives.
From voice to facial recognition, machines are learning about how humans use language and how they look. To do that requires the computational analysis of millions and trillions of data points, often in real-time, and of the datasets that data can be delineated into. Now imagine this technology applied to financial markets.
The networks of data points, on many related levels, pertaining to markets are neural in scale, requiring the analysis of millions of ‘moving parts’ in real-time and their timely interpretation. Investment banks are already using AI, ML and quantitative analysis to model what makes markets tick under varied conditions and circumstances, feeding in inputs from sectoral, regional, bottom-up stock analysis, macroeconomic and sentiment data, to name but a few variables.
This work is going on behind closed doors and beyond the reach of mere mortals such as private investors. However, much of the information and data is in the public domain and that which isn’t can be bought for a price.
What is needed is a platform that can pull that all together for the individual traders, so that each of us can access the computational power required to make sense of the data, that integrates prediction markets to harness the wisdom of crowds; bringing data scientists and crypto traders into each other’s orbits.
Essentially, this entails applying machine learning to already established indicators such as Bollinger bands, moving averages and Fibbonaci retracement patterns, and other indicators yet to be invented, such as from the world of social media-based (that’s where you come in) sentiment analysis or any other data feed or newsflow source a trader might want to build into their strategy.
Data science meets crypto trading
All this and more is being developed by Signals, a blockchain project based in the Czech Republic, whose mission is to make the development of machine learning-assisted trading and investment strategies an essential easy-to-use tool for crypto traders, and no doubt by later extension traders in mainstream markets such as equities, bonds and foreign exchange.
But first things first. Let’s look at how Signals is planning on making this work for the crypto trader. In what it calls ‘signal extraction’, its team of data scientists and algorithmic trading experts is making use of neural networks to discern patterns from technical and other indicators, including their relationships with each other as well as interaction with exogenous factors.
The models created from this process can then be back-tested on historical data, from which the models were originally developed. This would be an iterative process of continual testing beyond that automated in the machine learning.
And instead of trying to reinvent the wheel, Signals is accessing the power of other decentralised platforms, such as Gnosis and Stoxx in the predictions marketplace and the likes of SONM in distributed supercomputing.
Build your own strategy with drag and drop and sell your strategy
These powerful platforms and connected technologies can be interfaced with using the SGN token on the Signals Strategy builder, which in turn is plugged into the Signals Indicator marketplace ,where signals strategies can be bought, sold and followed in a social trading paradigm.
The Signals Strategy builder will let users drag and drop indicators and datasets on to a graphical canvas, along with conditional statements and operators in a way that is understandable to non-coders so that anyone can use it to build their own algorithmic trading strategy able to learn (machine learning) and react (artificial intelligence).
A by-product of this approach is the removal of cognitive bias from investment decision-making. Trading is an emotional business for most people, but it shouldn’t be because that’s when bad decisions are usually made.
Signals platform, which aims to be completed by the end of 2019 with crowd-sourced sentiment integrated into the live trading platform, will let anyone for minimal financial outlay access and develop sophisticated automated trading strategies to maximise profit and mitigate risk in a transparent environment.
Signals’ product is entirely different to those initial coin offering propositions that are merely hawking their own algo trading platform as opposed to letting a trader or investor develop their own.
Signals’ approach means, going forward, that the insights of many as opposed to just one small team, will be fed into a marketplace where potentially everyone learns from everyone else by seeing what strategies are the most profitable in a dynamic ecosystem, in which strategies are constantly being trained.
The future of trading
That represents an exciting step change, which represents a qualitative leap forward for trading techniques and strategy because it takes us in a direction that improves the profitably of trading.
It removes gut instinct from the equation and the tendency of non-professional traders, who often lack the discipline required to stick to pre-determined trading rules, especially in volatile markets such as cryptocurrencies, to stick with a strategy.
Instead, through the application of clearly developed and understandable rules and indicators in specific testable configurations and relationships, Signals’ trading platform strips out human bias and brings programmable tools to the lay person.
Ordinary traders will be able to develop and test their hunches about the significance of different metrics, indicators and sentiment information, and the patterns they throw up, by interpreting programmatically and artificially intelligently to achieve predictable results (profits) from a multitude of inputs.
Signals is building a profit engine and you may wish to join them on what could be a profitable journey into the future of investing and trading.