Articles tagged by machine learning
The current buzzwords are “AI” and ‘machine learning” – barely a day goes by without receiving a missive about yet another “ground-breaking” initiative, most of which appear to be nothing new, it’s just the PR managed to crowbar the words in there. Without doubt, though, firms are looking at this technology and, in some cases actually deploying it, which begs the question – who do we blame if the machine goes rogue? And who is the John Connor of FX?
The Financial Stability Board (FSB) has published a report that considers the financial stability implications of the growing use of artificial intelligence (AI) and machine learning in financial services.
It notes that financial institutions are increasingly using AI and machine learning in a range of applications across the financial system including to assess credit quality, to price and market insurance contracts and to automate client interactions. The lack of interpretability or auditability of AI and machine learning methods could become a macro-level risk, FSB warns.
Despite the hype around artificial intelligence and machine learning in an increasingly data-driven environment, Galen Stops finds that humans remain a vital part of the trading process.
Intel co-founder Gordon Moore famously noticed that the number of transistors per square inch on integrated circuits had doubled every year since their invention. This observation, which has become known as Moore’s Law, essentially predicts that this trend will continue into the foreseeable future, meaning that computing power will become more and more efficient.
Likewise, the acceleration of technology in financial markets – including FX – has meant that these markets have become increasingly efficient.
Last month I wrote about the challenges of regulating machine learning, but will AI highlight the different market structure between equities and FX – something that is a long running theme of this column? The value of AI is unarguable, but it strikes me that it will be put to different uses in FX than, for example, equities - and that is because of the different market structures of each instrument. One use is revolutionary, the other? Well we're kind of used to it...
Although fintech solutions are likely to change how FX operates throughout the trade lifecycle, expect these changes to be evolutionary rather than revolutionary, explained speakers during a recent Profit & Loss webinar.
The word “disruption” has become synonymous with fintech in recent years, with numerous articles, whitepapers and analyst reports warning that fintech upstarts are looking to upset the applecart in financial services.
Yet speakers on a recent Profit & Loss webinar, FinTech in FX: Getting Beyond the Hype, which was sponsored by IHS Markit, preferred to talk in terms of innovation rather than disruption when discussing the impact of fintech in the FX markets.
You can’t fight progress, but you can rein it in and make sure it goes in the right direction – advances are not always positive. There is so much chatter about financial markets withering and dying if they do not go the fully quantitative path, but is that right? I understand that these firms are largely hiring engineers and mathematics or physics grads but while these people have undoubted strengths and can seriously add value to a business, they are not the be-all-and-end-all.
There is a lot of conversation around Artificial Intelligence (AI) among different participants in the institutional investing pyramid.
Investors are wondering if AI can get higher returns by extracting unexplored alphas or if it can reduce costs, and investment professionals are wondering how machine learning and AI will impact their businesses.
Right now there is a lot of exuberance, optimism, skepticism and fear around AI and the impact that it will have on financial markets. Here I explore five key questions that are important to ask regarding this technology and its role in finance.
Hasan Amjad, head of algorithmic trading at GAM Systematic Cantab, explains how machine learning tools and techniques have enabled his firm to improve almost every aspect of its trading capabilities.
“It goes all the way really,” he says, “Starting with portfolio construction, all the way to the final trade and the post-trade analytics.”
For example, Amjad points out that machine learning can be used to improve pre-trade analytics by more effectively identifying what kind of trading the firm should be engaging in during current market conditions. He concedes that there are other techniques that enable firms to determine market conditions, but that “machine learning just takes it that one step further by being able to ingest a lot more data and give you the answer”.
SmartStream Technologies has launched a new innovation team tasked with creating solutions using artificial intelligence (AI), machine learning (ML) and blockchain technologies, in the areas of reconciliations, cash management, and fees and expense management.
“Highly skilled members of the team include mathematicians, applied data scientists, computer scientists and PhDs, who will focus on the deployment of AI/ML and blockchain models with financial institutions. This includes evaluating optimal AI/ML modelling, data interpolation, running tests, implementations and analysing how AI processes best work within the current product environment by monitoring achievements and optimisation of processes, to enable better business outcomes,” the firm says in a release issued today.
Charles Ellis, a trader and quantitative strategist at Mediolanum Asset Management, explains how data can be used to help generate alpha signals.
The first thing that Ellis points out is how trading firms can most effectively use data is dependent on their investment process and the type of research questions they are trying to use the data to answer.
For starters, he says, firms need to consider what investment time frame they are working towards.
“Then you have to ask which of these time frames can we add the most value to? What data do we have access to? And then it goes into what sort of questions can we answer using this data over these time frames?” comments Ellis.
One of the key benefits of the use of artificial intelligence (AI) tools for trading is that it can massively enhance human capabilities, explains Andrej Rusakov, CEO of Data Capital Management.
“The way I see it is that AI can really put human ingenuity on steroids,” he says. “What I mean by that is that it really allows you to take way more data points into account and find structures in data sources that are impossible for the human eye to spot.”
Rather than displacing humans, Rusakov explains that this technology is most effective when it is deployed in tandem with a human understanding of how markets work. When building strategies, his firm uses this understanding of markets and then codifies and enhances them by using AI, and in particular machine learning, tools to find new patterns in different data sets.
Online broker CMC Markets has partnered with Tradefeedr, a data science platform built for financial markets, to deploy cloud based machine learning to improve trading analytics and intelligence around liquidity management.
The firm says that the additional capabilities provide for the ingestion, cleansing and store of massive amounts of market and transactional data; high performance computing infrastructure for inspecting, intersecting and querying massive data sets, including data visualisation tools and APIs for extracting the results of analysis for further analysis.
Artificial intelligence (AI) and machine learning have become buzzwords in financial services, but while this technology can be applied in finance in numerous ways to improve returns, it also has some significant limitations that market participants should be aware of.
This was the message from speakers at the Profit & Loss Forex Network New York conference, on a panel discussion titled “AI: Regular Quants with a Bigger Bazooka?”
“In my mind the biggest problem with machine learning in its application to finance is the problem of non-stationarity.
Artificial intelligence (AI) and machine learning (ML) are reshaping the alternative investments landscape, but professional financial managers still make the most pivotal decisions, according to a new survey from BarclayHedge.
In a sample of 55 hedge funds that responded to the survey, 56% said they use AI/ML to inform investment decisions, with most of the firms that use these tools saying that they do so in order to generate trading ideas and optimise portfolios.
Well over half of the respondents, 58%, have used AI for three or more years, while 37% have used the technology for five-plus years.
Hedge fund managers were among the earliest adopters of advanced algorithms and artificial intelligence techniques, which helps explain why a plurality of survey respondents said they have been using AI/ML for more than five years.
FICC data analytics company Mosaic Smart Data has launched a new feature for its MSX platform enabling users to instantly generate text reports on their trading activity data using machine learning.
The feature will be available to all MSX users and will allow a trading activity report, which would take a member of staff hours to create, to be generated instantly.
The firm says the new service uses a machine learning technique called natural language generation (NLG), meaning MSX can generate trading activity reports on any set of analytics on the platform including both voice and electronic trade data.