Artificial Intelligence and Machine Learning (AI & ML) and Sentiment Analysis are said to “predict the future through analysing the past” – the Holy Grail of the finance sector. They can replicate cognitive decisions made by humans yet avoid the behavioural biases inherent in humans.
Processing news data and social media data and classifying (market) sentiment and how it impacts Financial Markets is a growing area of research. The field has recently progressed further with many new “alternative” data sources, such as email receipts, credit/debit card transactions, weather, geo-location, satellite data, Twitter, Micro-blogs and search engine results. AI & ML are gaining adoption in the financial services industry especially in the context of compliance, investment decisions and risk management.
This is a sophisticated conference that not only interrogates and explores the implications of AI & ML in the financial services industry but also goes on to identify the investment opportunities of sharing knowledge and exploiting IP in the finance domain.
Attend this event and earn GARP/CPD credit hours.
UNICOM has registered this program with GARP for Continuing Professional Development (CPD) credits. Attending this program qualifies for 7 credit hours. If you are a Certified ERP® or FRM®, please record this activity in your Credit Tracker.
We are inviting speakers – thought leaders, subject experts and start up entrepreneurs – to share their knowledge and enthusiasm about their work and their vision in the field of AI, Machine Learning, Sentiment Analysis.
Please complete the speaker’s response form and submit a proposal to present at this event.
UNICOM’s Code of Conduct & Views on Diversity
We at UNICOM strive to be a leading provider of knowledge to the business community and to engage the global business community as a specialised provider of knowledge. We strive to do this maintaining a culture of co-operation, commitment and trust. We want every UNICOM conference and training day to be a safe and productive environment for everyone – a place to share research and innovation and to build professional networks. To that end, we will enforce a code of conduct throughout all our events. We expect cooperation from all participants to help ensure a safe environment for everybody.
Our approach is that our events are dedicated to providing a harassment-free experience for everyone, regardless of gender, gender identity and expression, age, sexual orientation, disability, physical appearance, body size, race, ethnicity or religion. We do not tolerate intimidation, stalking, harassing photography or recording, sustained disruption of sessions or events, and unwelcome physical contact or sexual attention. We do not tolerate harassment of conference participants in any form. Sexual language and imagery is not appropriate for any conference venue, including talks, workshops, Twitter and other online media. Event participants violating these rules may be sanctioned or expelled from the event without a refund at the discretion of the conference organisers. Please bring your concerns to the immediate attention of the event staff.
Diversity: In our endeavour to be the provider of knowledge to the business community, we understand that this depends on hearing from and listening to a variety of perspectives that come from people of all races, ethnicities, genders, ages, abilities, religions, sexual orientation, and military service. We welcome diverse speakers for all our events, we do not always fully achieve this goal, but it is an ongoing process.
Ronald Hochreiter, Docent & CEO, WU Vienna University of Economics and Business & Academy of Data Science in Finance
AI and Machine Learning methods can be used to generate investment decisions successfully. A clever combination of Data Science methods with methods from the field of Decision Science (Prescriptive Analytics) may lead to even more successful models. In this talk a general outline for such a successful methodological combination will be presented as well as a concrete novel Deep Learning investment model which is based on graphical TTR series representations instead of using time-series directly. It will be shown how important Feature Engineering for Deep Learning in Finance actually is.
Enza Messina, University of Milano-Bicocca
We analyze how AI and Machine Learning and Sentiment Analysis of News and Micro-blogs are and impacting the two rapidly expanding markets, namely, Financial market and Retail market. We support our analysis by a few Use Cases for these markets.
Gautam Mitra, CEO, OptiRisk System & Visiting Professor, UCL
We compute daily trade schedules using a time series of historical equity price data and applying the powerful mathematical concept of Stochastic Dominance. In contrast to classical mean-variance method this approach improves the tail risk as well as the upside of the return. In our recent research we have introduced and combined market sentiment indicators and technical indicators to construct enhanced RSI and momentum filters. These filters restrict the choice of asset universe for trading. Consistent performance improvement achieved in back-testing vindicates our approach.
Claus Huber, Portfolio Manager, Deka Investment
A very valuable feature of the Self-Organising Map, a method of Machine Learning, is its visualisation capabilities. We show how the Self-Organising Map can be deployed to visualise the risk structure in a portfolio, in particular for assets for which no risk models exist. Some examples to this end are the visualisation of risk concentrations, identifying diversifiers and scenario analysis. Real-world applications are the selection of hedge fund managers or the analysis of a portfolio of Alternative Risk Premia.
Christopher Kantos, Senior Equity Risk Analyst, Northfield
In December of 2017 Northfield introduced the first commercially available factor risk models that incorporates computerized analysis of news text directly into volatility risk forecasts for individual stocks, corporate bonds, industry groups and ETFs based on market indices. Market events in early 2018 provided several excellent examples of why we believe that Risk Systems That Read® is the most significant innovation in factor risk models in more than three decades. We will illustrate show how recent news events drove financial market outcomes for Wynn Resorts, Wynn Macau, Facebook and Wanda Hotels (HK). Each day the content of thousands of news articles are now part of the input for the full range of models available from Northfield. The line of research that led to this innovation stretches back to 1997, and includes five published papers by Northfield staff [diBartolomeo and Warrick (2005), diBartolomeo, Mitra, Mitra (2009), diBartolomeo (2011,2013,2016)]. Beyond the obvious improvement in risk estimation, the method has important implications for alpha generation by both quant and traditional for active managers.
Lucas Bruggeman, Managing Director Sentifi AG
Finding an edge is hard because traditional data sources are widely available, have time lags and in some instances are manually aggregated. In addition to that, asset valuations shift constantly, and those who have the earliest information signals benefit the most. We address these challenges by commingling traditional and alternative data - enabling investors to identify market outliers and early indicators for future price movements.
This talk will answer the following questions:
♦ How significant is a news event for my portfolio?
♦ Is momentum shifting for stocks in my portfolio?
♦ Which companies are affected by risk events?
Panos Parpas, Senior Lecturer, Dept. of Computing, Imperial College London
Breakthroughs in modern Neural Network (NN) architectures and related algorithms in Machine Learning (ML) have entirely transformed whole areas of computer science such as computer vision and natural language processing. Unfortunately, both theoretical and empirical results have shown that neural networks compute unstable classifiers. An unstable classifier is vulnerable to adversarial attacks and illegal exploitation. The perturbations needed to fool ML classifiers are small, indistinguishable from noise, and therefore, they are difficult to detect. A necessary condition for successful ML systems in real-world applications is that the underlying system is stable. Without resolving this challenging problem, it is not possible to make meaningful progress in critical application areas such as the explainability and interpretability of machine learning algorithms, or efficient and robust training methods for reinforcement learning. Despite several attempts to address this problem, it remains open. In this talk, I review some recent developments that attempt to address this problem.
Matthias Uhl, Executive Director in Analytics & Quantitative Modelling at UBS Asset Management
We show that sentiment from news articles can explain and predict movements in the term structure of U.S. government bonds. This effect is stronger at the short end of the curve, coinciding with greater volatility and investors’ need to continually reassess the Fed’s reaction function. Facing such uncertainty, market participants rely on news and sentiment as a central element in their decision-making process. Considering this dependence, we propose a new yield curve factor-news sentiment-that is distinct from the 3 established yield curve factors (level, slope, and curvature) as well as from fundamental macroeconomic variables.
Francesco Cricchio, Co-founder and CEO of Brain
Brain developed a sentiment Indicator on about 7000 global stocks monitoring the news from financial media with the objective of providing the ‘market mood’ on listed companies.
The proprietary sentiment scoring indicator (BSI) is based on machine learning techniques used to categorise the news in specific topics combined with natural language processing techniques that make use of a proprietary dictionary of financial terms.
The sentiment indicator can also be aggregated on thematic baskets of stocks related to non-conventional themes (e.g. nanotechnology, wearables, drones) selected by machine learning analysis of company public information (e.g web pages) of global stocks.
Richard Peterson, Managing Director, MarketPsych
In this talk Dr. Richard Peterson describes how media analytics are providing new insights into the origins and topping process of asset price bubbles. Examples from price bubbles including the China Composite, cryptocurrencies, housing, and many others will be explored. Recent mathematical models of bubble price action will be augmented with sentiment analysis. Attendees will leave with new models for identifying and taking advantage of speculative manias and panics.
Dr. Jochen L. Leidner, Director of Research, R&D, Refinitiv Labs, Refinitiv Ltd., London; Royal Academy of Engineering Visiting Professor of Data Analytics, University of Sheffield
It has long been stated that sentiment is an important element of financial markets. But what exactly is sentiment, and how does it related to risk? In this talk, I will outline uses of sentiment that turn out to be fundamentally different, and contrast it with the more objective notion of risk. I point out some flaws in past work on sentiment and describe how they can be avoided. Finally, I relate sentiment, risk and ESG to each other: recently there has been increased attention in finance to supplement fundamental analysis of companies with non-financial metrics to address the desire to account for Environment, Sustainability and Governance (ESG) factors also known as Corporate Sustainability (CSR) for short). I argue that by exploring sentiment in an fine-grained, aspect-based way, we can derive more valuable insights.
Kamilla Kasymova, Associate, Quantitative Research and Analytics, Northwestern Mutual and Jacob Gelfand, CFA, Director of Quantitative Strategy and Research, Investment Risk Management
We present a framework for ex-ante analysis of the USD denominated EM Sovereign spreads and respective currencies based on the flow of news in global and local media. The public domain GDELT data is currently being used, but the framework is agnostic to data source, and can be adopted to any data source. We will discuss use cases pertaining to selected countries and spearhead the discussion about predictive qualities of the produced analytics.
Boryana Racheva-Iotova, Head of Risk and Quantitative Analytics | FactSet
We explain main concepts of Prospect Theory (PT) and Cumulative Prospect Theory (CPT) within the framework of the rational dynamic asset pricing theory. We discuss key requirements with respect to the CPT’s weighting function to preserve key characteristics of the price process essential for deducting derivatives pricing asset and solve general class of optimization problems. We provide an example when the asset returns are altered with a a modified Prelec’s weighting probability function. We introduce new parametric classes for Prospect Theory value functions and weighting probability functions consistent with rational dynamic pricing Theory and illustrate the concept by deriving market-sentiment measures based the options market. Practitioners could use such sentiment measures as new type of risk factors. The approach can also be used in more general optimization problem as a substitute for the empirical density as it derives the entire market-implied distribution.
Lily Yingyi Gu, Quant Reseacher, Bloomberg LP
We model the volatility surface dynamics with Multilinear Principal Component Analysis (MPCA), a novel and natural adaptation of PCA on tensor objects. We illustrate this approach's robustness to small sample size and how it improves the interpretability of the resulting eigen-surfaces compared to PCA.