Trading Fast And Slow

I Know First Research Team LogoThis article was written by the I Know First Research Team.

It would be an obvious truism to say that the pace of life is increasing by day. Faster travel with high-speed rail and planes, faster communication with new and new generations of wireless networks, faster data-crunching enabled by top-notch hardware… “Faster” is also the keyword when it comes to trading – or, at least, it used to be, because the appetite for high frequency trading is seemingly lowering. Is it time to push the brakes instead of putting the pedal to the metal? What are the respective benefits of trading fast and slow, and which one is better for you? Keep reading to find out.

Gotta Go Fast: Rise And Decline Of HFT

When it comes to trading fast, high-frequency trading, as you can tell by the name, is pretty much the definition of it. This approach, as we discussed earlier, sees computers follow complex algorithms as they conduct dozens of gargantuan transactions every second. For the traders, this means accumulating small, even minuscule, margins into large gains, and for the trading platform, this means an infusion of liquidity. The latter often allows HF-traders to save on dragdown expenses like fees or commissions.     

(Source: Pixabay.com)

Ultra high-frequency traders double down on this by gaining access to exchanges that post stock price updates before they are available to everyone. The core of the practice, however, remains the same: you react before the market has had any time to react and jump at any temporary fluctuations resulting from the actions of large players to gain profit. Long-term trends are not what you are in for; you live and trade in the moment, closing all of your positions by the end of the trading session.

Arbitrage is the most likely source of your profits as an HF-trader, whether you are operating within one market or have an operation that stretches across multiple exchanges and asset universes. But while all this has probably sounded lucrative beyond belief, the reality is that HFT is less and less attractive these days, even though it would probably still be wrong to say its outright dead.

The first problem is that an HFT operation comes with a lot of expenses. The huge data centers filled to the brim with powerful computers crunching the numbers as they execute the complex algorithms take a lot of power to run. That is the smaller of the issues, though, as the competition puts HF-traders under a lot of pressure to constantly upgrade both the hardware and software they are using, which also leads to enormous bills. This means that HFT as such is a domain of large institutional investors, which fuels the controversies around the strategy. Add to this the increasing government scrutiny, and you have a whole load of issues biting into your profit margins – and those were small to begin with, remember?

This is why trading fast is not an option for private investors, and for smaller institutionals as well. What are the other options on the table? Let’s look.

Slow And Steady Wins The Race

As we discussed, HFT comes equipped with a whole variety of drawbacks that makes it less advisable to large players these days, and glaringly inefficient for small and retail traders. Trading strategies operating within less pressing time-horizons have their drawbacks as well, but they do not require you to run a sophisticated data-center to work.

(Source: Pixabay.com)

Here, the key question to ask yourself is how slow you want to go. If you are still inclined to go fast, just not as fast as computers, you can try to profit off the daily price fluctuations and go for intraday trading. There is more than one strategy featuring this kind of approach, including the HFT. We can also point at scalping, however, which also seeks to gain from daily price fluctuations, relying on technical analysis to pick the stocks to long and short. It also operates on large numbers to make up for the small margins, which, once again, is suboptimal for smaller players. News trading is another option, which sees the trader monitor the news flow and respond to the stories that can impact the market.

As you have noticed, all of these strategies ultimately seek to make use of short-term price fluctuations, or swings, which is why we can refer to them as swing trading strategies. Swing traders do not have to close all of their positions by the end of the day, of course: other time frames are also feasible with this strategy. If you foresee a price fluctuation that stretches beyond the next few minutes, you can buy and hold a stock for a few days or even weeks to make the most out of its rise. The same applies to shorting stocks that you expect to go down.

Alternatively, you can rely on the so-called position trading. Here, what you are looking for is not a temporary fluctuation, but rather a trend in the movement of a specific stock. The latter is, once again, mainly identified by means of technical analysis. Instead of exiting quickly, a position trader holds the stock for a few months to make the most out of the trend before the tide changes. Thus, as far as speculative trading goes, as opposed to long-term investment, this approach is one of the slowest ones.

Slower strategies typically require less investment into sophisticated high-tech hardware and imply less of a need to be constantly hooked on the markets. What is even more interesting for traders, they can also benefit from high-tech solutions without getting as pricey as HFT. Just as many others, slower traders stand to greatly benefit from the era of AI descending upon us.

Trading Fast And Slow With AI

The artificial intelligence industry is blossoming these days, transforming the world. With advanced statistical programming at their core, artificial intelligence algorithms excel at working with large datasets and picking up trends and patterns. This makes them a perfect fit for technical analysis and the trading strategies that are based on it.

(Source: Pixabay.com)

AI can identify and predict the longer-term direction of the market and the strength of trend. Knowing that, one can use it for getting the best intra-day entry points in the market, which can get pretty chaotic every now and then.

News trading may be a better fit for AI to shine, since natural language programming at its current stage can easily help the AI pick up the trading signals in the flow of textual information, whether it is about financial statements or an actual news feed. But when it comes to technical analysis, AI would make for a great assistant to those opting for longer-term swing strategies and position trading. AI predictions can be used to find both the optimal assets to go for and the best timing to exit, making sure the trader has made the most out of the transaction.

Such predictive algorithms are already there, and not just locked up behind the closed doors of large institutional players, like JPMorgan Chase. Tech-savvy startups are out there, looking out for the smaller players seeking to level the playing field. Among them is I Know First, an Israeli company that has trained its own stock market predictions AI.

The deep learning algorithm delivers daily forecasts for over 13,500 financial instruments, including currencies, ETFs and stocks. Trained on a set comprised of 15 years-worth of trading data, it views markets from a holistic perspective, processing fresh market data to model the trends and pick out the best stocks to long and short.

(Source: Iknowfirst.com)

Its forecasts are presented as a heatmap with two numeric indicators: signal and predictability. Signal works as a measure for the performance of a given asset against the rest of the financial instruments on the forecast. A strong positive signal means the asset is expected to rocket, while a strong negative one is a promise of a nosedive. Predictability shows how accurately the algorithm has been predicting the asset before. It ranges from -1 to 1 and is defined as the Pearson correlation rate between earlier forecasts and actual price movements.

The algorithm can adapt to virtually any market conditions, because it incorporates elements of genetic coding. In other words, it is aware of its own performance and updates its predictive models when they seem to be getting out of touch with the market. Accordingly, its accuracy goes up after each learning cycle.

The AI also draws on chaos theory to account for market volatility. It delivers its forecasts for time horizons ranging from 3 to 365 days, covering short, medium and long-term perspectives. This leaves intraday trading outside the current range of its possibilities, but is useful for everyone on the slower side of things. In a string of recent evaluations like this one, it demonstrated an accuracy of around 60% for 3-day predictions, rising up to 80-90% for 3-month forecasts. Thus, both longer-term swing traders and position traders are set to benefit from its predictions, using them to pick up the best stocks to long and short.