JPMorgan's guide to quantum machine learning in finance
We suggested in January that it might be a good idea to familiarize yourself with quantum computing if you want to maximize your future employability in financial services. A new academic paper from JPMorgan's Future Lab for Applied Research and Engineering helps explain why.
Authored by Marco Pistoia, JPMorgan's head of quantum technology and head of research, plus members of his team, the paper stresses that quantum computing will impact financial services sooner than you think. Goldman Sachs and JPMorgan have both been building teams of quantum researchers and Goldman has already used quantum methods to speed up derivatives pricing by over a thousand times. The finance industry stands to benefit from quantum computing "even in the short term," says JPMorgan.
The researchers note banks and finance firms are already big users of machine learning techniques like reinforcement learning for algorithmic trading, or Natural Language Processing (NLP) for risk assessment, financial forecasting and accounting and auditing. Many of the machine learning techniques using quantum methodologies, but talent remains hard to find. "Demand is high and quantum is still a very rare skill," says one senior banking technologist.
1. Asset pricing
Banks have been using Recurrent Neural Networks (RNNs) to run time series predictions and are considering using them for asset pricing models, says JPMorgan. However, RNNs consume a lot of computing power, and there are advantages to using parameterized quantum circuits (PQCs) and quantum Long Short Term Memory (LSTM) units that allow users to make predictions about evolving processes from historical data.
2. Predicting volatility
Quantum methods can also be used to determine the likely changes in a security's price. Deep quantum neural networks produce a density matrix, and the implied volatility of an option is calculated using its respective element in the matrix.
3. Predicting the outcome of exotic options
Machine learning support Vector Machines (SVMs) can be used to predict the circumstances in which exotic options used in markets like FX pay out. Quantum techniques can facilitate this.
4. Fraud detection
Quantum clustering algorithms can be used to perform anomaly detection and identify fraudulent activity.
5. Stock selection
Quantum clustering algorithms can also be used to cluster stocks with similar returns but different risks, thereby allowing investors to pick low-risk stocks with high returns.
6. Hedge fund selection
The same clustering algorithms can be used to identify hedge funds for funds to invest in based on known variables like asset classes, size, fees, leverage and liquidity.
7. Algorithmic trading
Quantum reinforcement learning techniques could be applied to algorithmic trading in order to speed up decision-making and improve the complexity of models. However, the researches note that this hasn't happened yet due to the hardware limitations of current quantum devices.
8. Market making
Electronic market makers like Citadel Securities and Jane Street are likely to take an interest in quantum computing for their own reasons. "Market making is amenable to quantum reinforcement learning," says JPMorgan. - The problem is modelled as an "agent state, taking into account attributes such as inventory and risk-tolerance, and an environment state where the agent only has partial information."
9. Financial forecasting, accounting and auditing and risk assessment
JPMorgan's team predict that quantum natural language processing (NLP) algorithms are also coming for jobs in risk and accounting teams. NLP can be used, for example, to elicit "lender’s and borrower’s emotions during a loan process, to conduct sentiment analysis for forecasting, or to create semantic knowledge bases for financial accounting standards.
Contact: firstname.lastname@example.org in the first instance. Whatsapp/Signal/Telegram also available (Telegram: @SarahButcher)
Bear with us if you leave a comment at the bottom of this article: all our comments are moderated by human beings. Sometimes these humans might be asleep, or away from their desks, so it may take a while for your comment to appear. Eventually it will – unless it’s offensive or libelous (in which case it won’t.)