Machine Learning for Trading

How AI helps traders make better decisions & improve high-frequency trading

Summary

Trading is a gruesomely competitive world. And with AI being painted as the new wonderweapon for everything, it’s understandable that there’s a huge amount of interest in discovering how to use AI for trading.

Disclaimer

AI does play an important role in trading – but maybe not in the way you’d expect. Unfortunately, AI can’t be used to power a superhuman trading machine that steals human traders’ lunches in every market. Not even Renaissance Technologies has that capability. At least not yet.

Why not?

The short answer: human competition – more about that below.

Meanwhile, the battles AI actually wins are much more incremental – but still significant. AI is more a trader’s side-kick than their replacement.

Below are the top four ways AI applies to trading today:

Analysing Sentiment

Humans can’t process all the information out there – but machines can get close. With AI, machines can now do much more – including analysing and summarising texts.

They can summarise things like sentiment: scrapers collect daily news, tweets, and other social media posts about a particular topic, and then AI algorithms (in particular Natural Language Processing) summarise whether positive or negative views are stronger. They can even categorise texts into topics and build human-readable text summaries automatically.

This is highly valuable information for traders, who have to be up to date on what’s happening as quickly as possible.

Forecasting real-world data

Traders also use AI to improve the reliability of input data forecasts – elements in the real world that help traders succeed.

Like the prediction of:

  • Weather patterns in the northeast over the next two weeks;
  • Solar energy supply in Europe;
  • The outcome of a political election.

These forecasts are based on other algorithms developed by other companies – but that doesn’t mean these predictions can’t be improved.

One handy trick is to train an algorithm to combine multiple expert forecasts into another prediction – which is then more accurate than any of the forecasts it was based on. This is called ensembling, and it works pretty well.

Finding patterns

Trading is about identifying localised patterns – which are often limited in time and space – and then guessing how to exploit them. The process of finding patterns is arduous and time consuming.

But AI algorithms are basically pattern-finding machines. If an analyst suspects irregularities in a particular dataset, they can save time by using AI to find them.

So AI can find useful patterns, as long as it’s guided by an experienced analyst who knows what to look for. These patterns are then used by traders, who mix them with their experience and intuition, then apply them. Or you can use them to design automated trading machines – see the next section.

Tuning high-frequency trading machines

In high-frequency trading – as the name suggests – machines execute thousands or millions of trades per day, trying to take advantage of inefficiencies that only exist over very short time spans.

Human beings can’t make these trades – there are simply too many – but humans define the rules by which these machines operate.

However, because the market keeps changing, these machines need to be adjusted constantly. That takes a lot of time and effort.

AI can automate these recalibrations – and do a lot of the repetitive statistical work that analysts would otherwise need to do.

Warning signs – Things to watch out for

The promise of finding a miracle algorithm that can really print money is so alluring, lots of smart people have bought into it. Here are the most important traps to watch out for:

AI today is NOT “smarter” than human beings

In truth, even today’s most advanced AI algorithms are very naive compared to human brains.

When an algorithm beats a human at chess or GO, it’s like a car beating a human runner in a quarter-mile race: yes, the machine is faster, but that doesn’t make it superior. It just means that we built a machine that can do a very narrowly defined task – under certain narrowly defined conditions – really well.

It might seem like it at first, but trading is not such a narrowly defined task. Why? Because in trading you are competing with other humans – who will use all their brainpower trying to outsmart you.

Published trading strategies often don’t work in real life

There is a lot of research and lots of blog articles out there promising a profitable AI-based trading algorithm. But these models don’t tend to work in real life, for several reasons.

Faulty setups

A surprising number of papers actually make mistakes in how they set up their training and testing framework. They use variables that wouldn’t have been available at the time when the AI needed to make a decision (data leakage) or evaluate the predictions against the current price, not the future price. Surprising mistakes – but time-series datasets are a complicted thing for human brains to handle.

Selection bias

Fund managers are often critiqued for attributing their better returns to superior skill rather than luck. But if a lot of fund managers make a lot of random guesses, then in the end there will be some who made a few good guesses. The losers close down, and we never hear about them – so it looks like there are lots of fund managers who have the skills to beat the market.

In reality though, the number of fund managers who beat the market is exactly in line with what you would expect based on random guesses.

The same is true for research papers. If you try a lot of algorithms, you’ll eventually find one that seems to produce certain profits. If you don’t tell everyone how many experiments you had to run to get there, it will seem like you just stumbled on a superior approach.

However, there is absolutely no guarantee that this strategy will work outside the particular data you tested it on.

Transaction fees and slippage

Actually, building a trading strategy that outperforms the market is often quite simple – IF you forget about the real-world costs of doing trades. Transaction fees (the fees you pay for every trade) and slippage (the fact that the price might change between the time you make your order and the trade going through) eat up a lot of profit. And in almost every case, that’s enough to delete the profit you saw in simulation.

Patterns change over time

One of the most important concepts in Machine Learning is finding patterns in past data and using them to make correct forecasts of the future.

However, this doesn’t work in trading. Other traders are competing to find the same patterns – so patterns get found, exploited, and then disappear. That means patterns rarely exist for long, and you have to constantly find new ones.

This requires immense adaptability – something human beings are currently much better at than machines.

Algorithms alone will never give you an advantage

It’s easy to get carried away and focus on the algorithm as the main competitive advantage between one trading strategy and another.

This is essentially what companies like Numerai are proposing:

  1. Combine lots of good models into a super model.
  2. Beat the stock market.

But this does not work. Why? Because data beats algorithms. The data you give to your algorithm has a much larger effect on your model’s performance than how good the algorithm is. The data Numerai gives you is fixed – you can’t add to it. So the forecasts they make will always be worse than those of traders, who aren’t limited in the data they can use – traders who have access to an open pool of data, who can try, test, and add new data points to their algorithms continuously.

A bit of hope - Inefficent markets might still be fertile ground

Markets with few trading participants, high barriers to entry, a limited trading volume, and few players able to use Machine Learning might offer some opportunity for pure-play AI trading success.

In these markets, automated trading – especially the use of Machine Learning – is still just beginning, and traders who build automated trading engines could score enough of an edge to produce a good profit.

Energy Transmission

Anticipating and Preventing Power Grid Failure

Massive power outages cause chaos for the general public, and they cost utility providers roughly $49 billion a year.

This wouldn’t be much of a problem if massive power outages were rare, but outages affecting more than 50,000 people have increased dramatically in recent years. This means utility companies need to find new ways of anticipating and managing these outages.

These days, smart grids are producing massive amounts of data, which means predicting and managing outages is easier than ever. Unlike traditional power grids, which are one-directional (meaning they only transmit power in one direction), smart grids are two-directional. They can capture data from every possible source in the grid at the same time as they’re providing electricity. They collect and monitor data from sources like smart meters, IoT devices, and power generation stations, providing a clear, real-time look at power usage.

Machine learning can use this data to anticipate and prevent massive power outages in the grid. Machine learning helps identify non-obvious patterns in the data that can be a precursor to grid failure, which helps maintenance teams preempt failure.

Balancing the Grid

Balancing the grid — making sure energy supply matches energy demand — is one of the most important jobs a transmission operator has. But renewable energy sources depend heavily on the weather, making them harder to predict.

Transmission operators spend millions each year fixing planning mistakes that lead to producing too much or too little power. In hybrid systems — which rely on both renewable energy sources and fossil fuels to generate electricity — these mistakes have to be corrected at the last minute by buying more energy or compensating power plants for the excess.

Machine learning is the most accurate method available to forecast the output of renewable energy. Advanced methods, like Long Short-Term Neural Networks (LSTMs), can weigh the many factors involved — wind, temperature, sunlight, and humidity forecasts — and make the best predictions. This saves money for operators and preserves resources for power plants.

Preventing Blackouts and Brownouts With Real-time Monitoring and AI Prediction

Power grids have a lot of obstacles to overcome in providing continuous energy to customers. Weather patterns, usage, internal failure, even wildcard incidents like lightning strikes and interference from wild animals can all affect power delivery.

Machine learning is increasingly being used to help predict potential brownout and blackout conditions. By feeding historical data into the AI and running Monte Carlo simulations to predict potential outcomes, grid operators can use machine learning to identify conditions that could lead to grid failure. And they can act accordingly.

Sensors like phase measurement units (PMU) and smart meters can provide usage information in real-time. When combined with both historical and simulation data, AI can help mitigate potential grid failure, using techniques like grid balancing and demand response optimization. Incidents that would otherwise have affected millions of people can be contained to a smaller area and fixed faster for less money.

Differentiate Power System Disturbances from Cyber Attacks

Cyber attacks are increasingly used to target important infrastructure, like shutting down hospitals with Ransomware attacks (when attackers break into the system and lock legitimate users out until a ransom is paid). With utility grids, a cyber attack can have widespread consequences and affect millions of users.

Detecting these attacks is critical.

Developers are using machine learning to differentiate between a fault (a short-circuit, for example) or a disturbance (such as line maintenance) in the grid and an intelligent cyber attack (like a data injection).

Since deception is a huge component of these attacks, the model needs to be trained to look for suspicious activity – things like malicious code or bots – that get left behind after the deception has occurred.

One such method uses feature extraction with Symbolic Dynamic Filtering (an information theory-based pattern recognition tool) to discover causal interactions between the subsystems, without overburdening computer systems. In testing, it accurately detected 99% of cyber attacks, with a true-positive rate of 98% and a false-positive rate of less than 2%. This low false-positive rate is significant because false alarms are one of the biggest concerns in detecting cyber attacks.

Balance Supply and Demand

Utility providers are looking for ways to better predict power usage while maintaining maintaining energy supply at all times. This becomes critical when renewable power sources (like solar or wind) are introduced into the grid.

Because these renewable power sources rely on elements beyond human control (like the weather), utility providers know they can’t always rely on renewables for continuous production. Knowing precisely when demand levels will peak allows utility providers to connect to secondary power sources (like conventionally generated electricity) to bolster the available resources and ensure constant service provision.

More and more utility providers are turning to machine learning for help. We can feed historical data into machine learning algorithms -- like Support Vector Machines (SVM) -- to accurately forecast energy usage and ensure sufficient levels and constant supply.

Detect Power Grid Faults

Current methods for detecting faults in the grid consume a lot of unnecessary time and resources. This creates a situation where power transmission is interrupted and customers are without electricity while faults are first located, then fixed.  

Machine learning can find faults quickly and more accurately helping you minimize service interruption for your customers.. Support Vector Machines (SVM) are combined with Discrete Wavelet Transformation (DWT) to locate faults in the lines using a traveling wave-based location method.

When we apply  DWT (a form of numerical and functional analysis that captures both frequency and location information) to the transient voltage recorded on the transmission line, we can determine the location of the fault by calculating aerial and ground mode voltage wavelets. So far, this method has detected fault inception angles, fault locations, loading levels, and non-linear high-impedance faults for both aerial and underground transmission lines.

Detect Non-Technical Power Grid Losses

In the energy world, “non-technical losses” means energy theft or fraud from the system.

There are two common types of non-technical losses. The first is when a customer uses more energy than the meter reports. The second involves rogue connections stealing energy from paying customers. To pull off this theft or fraud, bad actors can bypass smart meters completely or insert chips into the system that change how meters track energy use. Meter readers can also be bribed to report lower numbers (though thanks to smart meters, this is increasingly hard to do).

Because these non-technical losses cost $96 billion annually, utility providers are turning to machine learning to combat the problem.

We can help utility providers mine historical customer data to discover irregularities that indicate theft or fraud. These can be things like unusual spikes in usage, differences between reported and actual usage, and even evidence of equipment tampering.

Energy Distribution

Better Predict Energy Demand

Accurately predicting customers’ energy needs is critical for any utility provider. To date, we haven’t found an adequate solution for bulk energy storage, which means energy needs to be transmitted and consumed almost as soon as it’s produced.

We're using machine learning to increase the accuracy of these predictions. Historical energy use data, weather forecasts, and the types of businesses or buildings operating on a given day all play a role in determining how much energy is used.

For example, a hot summer day mid-week means more energy usage because office buildings run air conditioning at a high capacity. Weather forecasts and historical data can help identify those patterns in time to prevent rolling blackouts caused by air conditioners in the summer.

Machine Learning finds complicated patterns in the various influencing factors (such as day, time, predicted wind and solar radiation, major sports events, past demand, mean demand, air temperature, moisture and pressure, wind direction, day of the week, etc.) to explain the development of demand. Because machine learning finds more intricate patterns, its predictions are more accurate. This means energy distributors can increase efficiency and decrease costs when they buy energy – without having to make expensive adjustments.

Energy Generation

Predict Turbine Malfunction

Wind is a great renewable energy source, but wind turbine maintenance is notoriously expensive. It accounts for up to 25% of the cost per kWh. And fixing problems after they occur can be even more expensive.

Machine learning can help you get ahead of this problem. The goal is to reduce maintenance costs by catching problems before the turbine malfunctions. This is particularly important when wind farms are located in hard-to-access places, such as the middle of the ocean, which makes repair costs even higher.

Real-time data gathered with Supervisory Control and Data Acquisition (SCADA) can help identify possible malfunctions in the system far enough in advance to prevent failure.

For example, data from sensors found within the turbines – such as oil, grease, and vibration sensors – have been used to train machine learning models to identify precursors to failure, such as low levels of lubricant.

This method can train machine learning models to predict failures up to 60 days in advance.

Consumption / Retail

Accurately Predict Energy Prices

As personal power generation (using solar or wind power) gets easier and cheaper, consumers and businesses are increasingly producing their own power.

Personal power generation allows people to make, consume, and store their own energy. Depending on where they live, they may even be able to sell surplus power back to the local power utility.

Machine learning can help find the best time to produce, store, or sell this energy. Ideally, energy should be consumed or stored when prices are low and sold back to the grid when prices are high.

By looking at historical data, usage trends, and weather forecasts, machine learning models have made accurate predictions on an hourly basis. People with personal and business energy generation systems can use these predictions to make strategic decisions about whether to use, store, or sell their energy.

For example, Adaptive Neural Fuzzy Inference System (ANFIS) has been used to predict short-term wind patterns for wind power generation. This allows producers to maximize energy production and sell it when energy prices are at their peak.

Reduce Customer Churn

In open energy markets, where customers have a choice of utility providers, understanding which customers are going to churn out can be critical. Churn rates, which is the percentage of customers who stop using your service in a year, can be as high as 25%. Being able to predict churn and stay ahead of it is essential to survival.

Machine learning is helping utility owners predict when a customer is getting ready to churn out. By using techniques such as Cross-industry Standard Process for Data Mining (CRISP-DM), AdaBoost, and Support Vector Machines, as well as historical usage data, utility providers can identify key indicators of whether or not a customer is going to churn. These indicators include things like customer satisfaction, employment status, energy consumption, home ownership or rental status. A change in any of these can indicate a customer is getting ready to terminate their service.

When these indicators are identified far enough in advance, it’s possible to avoid churn by working with customers to solve any problems they’re experiencing.

Energy Trading

Predict Energy Prices

Just like natural gas and oil, wholesale energy is a market commodity. So naturally it's important for traders to be aware of market fluctuations and pricing when it comes to buying and selling energy.

To help make sense of the massive amounts of data used to make trading decisions, traders are increasingly turning to machine learning.

A mix of statistical analysis and machine learning can help commodity traders make better predictions. Classical statistical analysis techniques like time series analysis, Seasonal Autoregressive Integrated Moving Average (SARIMA), and regression models are used to deal with the data. And machine learning makes connections between the various data points.

What’s more, machine learning trains itself to make increasingly accurate predictions using the constant flow of real-time data.

Keep reading

No items found.
No blog posts found.