Conference Day One: 16 September 2019

8:00 am - 8:50 am Welcoming tea, coffee and registration

KEYNOTES & OPENING PLENARY SESSIONS

8:50 am - 9:00 am Chair's opening remarks

William Kelly, Chief Executive Officer at CAIA

William Kelly

Chief Executive Officer
CAIA

9:00 am - 9:20 am Opening keynote: Can artificial intelligence be ethical? Some lessons from early adopters of ethical AI

Artificial Intelligence is here to stay and is rapidly being adopted in virtually every domain – from healthcare to finance and public safety and entertainment, to name a few.  However, most AI systems and ML models today are black-boxes that often function in oblique, invisible ways for both its developers and especially compliance staff and regulators.  This has led to Ethical and Regulatory concerns about how AI may harm the most vulnerable driven the need for building Trusted and Ethical Responsible AI systems — AI applications and models that are free from bias, transparent in their operations, and are able to reflect the core values and policies of the business and governments. This session will address the importance of Ethical AI and how it is being applied currently by businesses and governments.
Manoj Saxena, Chairman at AI Global

Manoj Saxena

Chairman
AI Global

Bias has been proven to become a factor in AI models and can have significant impact on results. What steps can you take to minimise or control bias
Anthony Ledford, Chief Scientist at Man AHL

Anthony Ledford

Chief Scientist
Man AHL

Pavan Arora, Chief AI Officer at Aramark

Pavan Arora

Chief AI Officer
Aramark

Miquel Noguer I Alonso, Co-Founder at Artificial Intelligence Finance Institute

Miquel Noguer I Alonso

Co-Founder
Artificial Intelligence Finance Institute

Gary Kazantsev, Head of Quant Technology Strategy, Office of the CTO at Bloomberg

Gary Kazantsev

Head of Quant Technology Strategy, Office of the CTO
Bloomberg

Mario Schroeck, Former Head of AI at Amplitude Capital

Mario Schroeck

Former Head of AI
Amplitude Capital

Trust and interpretability are vital in allowing asset allocators to increase investment in AI driven funds.  What are the latest developments, and how can you separate the claims from the facts.

Session preview:
Artificial intelligence (AI) has been increasingly hyped over the past few years. This is led to both great hope and great anxiety, primarily rooted in an assumption of how much the output of AI can be trusted. Here we examine some popular examples and posit a framework for how to incorporate AI into existing systems while trusting it neither too much nor too little. Our approach is geared towards utilizing AI for tasks where it is strong while implementing multiple controls that mitigate risk in regimes in which it is weak. Notably, this requires multiple touchpoints with experts in a partnership of human and machine. 
Afsheen Afshar, Former Chief Artificial Intelligence Officer and Senior Managing Director at Cerberus Capital Management

Afsheen Afshar

Former Chief Artificial Intelligence Officer and Senior Managing Director
Cerberus Capital Management

10:30 am - 10:50 am Networking refreshment break in the exhibition area

MORNING SESSIONS

How do you decide about how much to invest in AI driven funds, and justify the decision to your board and shareholders? A panel of leading asset allocators share their decision making process and discuss the importance of issues like interpretability and compliance
Afsheen Afshar, Former Chief Artificial Intelligence Officer at Cerberus Capital Management / JPMorgan Chase

Afsheen Afshar

Former Chief Artificial Intelligence Officer
Cerberus Capital Management / JPMorgan Chase

Irina Bogacheva, Head of Multi-Asset Research at QS Investors

Irina Bogacheva

Head of Multi-Asset Research
QS Investors

Fredrik Nerbrand, Former Multi-Asset Strategist at BlackRock

Fredrik Nerbrand

Former Multi-Asset Strategist
BlackRock

Joseph Simonian, Senior Investment Strategist at Acadian Asset Management

Joseph Simonian

Senior Investment Strategist
Acadian Asset Management

11:35 am - 12:05 pm The past and future of quantitative research

Traditionally, the development of investment strategies has required domain-specific knowledge and access to restricted datasets. The great majority of data scientists lack either or both of these requirements. The future of quantitative research passes through the design of platforms that enable all data scientists to identify market inefficiencies, hence increase market efficiency by democratizing finance.
Marcos López de Prado, Professor of Practice at Cornell University

Marcos López de Prado

Professor of Practice
Cornell University


Sandy Rattray, Chief Investment Officer at Man Group PLC

Sandy Rattray

Chief Investment Officer
Man Group PLC

Jonathan Guthrie, Head of Lex Column at Financial Times

Jonathan Guthrie

Head of Lex Column
Financial Times

12:35 pm - 1:30 pm Networking lunch in the exhibition area


Lunch and Learn

12:40 pm - 1:10 pm Alpha capture for alternative datasets

Twenty years ago fund managers asked if there was alpha in traditional research and stockbroker recommendations. What followed over the next two decades was the establishment of alpha capture networks to measure the value of broker recommendations. 
 
Today, with the exponential growth in alternative data, we ask ourselves a similar question – is there alpha in alternative data? To date this has been ascertained through extensive alpha testing by the buyers of alternative data. The recommendations and case studies from vendors are questioned like broker reports of 20 years ago.
 
At AI & Data Science in Trading, Eagle Alpha, is launching the first alpha capture solution based on alternative datasets. As the largest aggregator, with over 1,000 dataset relationships and growing, we are uniquely positioned to independently aggregate, track and measure dataset predictions. 
 
The key benefits for buyers of alternative datasets include dataset prioritization and capturing alpha. 
Emmett Kilduff, Founder at Eagle Alpha

Emmett Kilduff

Founder
Eagle Alpha

Ian McFarlane, Head of Client & Product Engagement at Eagle Alpha

Ian McFarlane

Head of Client & Product Engagement
Eagle Alpha

AFTERNOON STREAMS

STREAM A

1:30 pm - 1:35 pm ALT DATA

Emmett Kilduff, Founder at Eagle Alpha

Emmett Kilduff

Founder
Eagle Alpha

STREAM A

1:35 pm - 1:50 pm The appeal and perils of quant assisted social listening - alt data's ground zero for information discovery
Session preview:
The dissemination of socially sourced, non-financial information to Wall Street and the investing public is inefficient, leading to information imbalances that can last from seconds to weeks. As global mass adoption of social networks progresses the speed, reach, and mechanics of modern communication - the arc of data dissemination flattens. This evolution provides an opportunity for the interpretation of social metrics, cues, and warnings depicting the present in close to real time. 

However, to the degree that transactional, usage, and geolocation alt data sources are reveled for their precision - social data can be opaque, contextually noisy, and highly interpretable. The inherent hurdles and perceived weaknesses of social data present an opportunity for individual quanta mental traders to develop peerless and differentiated expertise in its measurement and interpretation. 

The discussion will dissect key differences between non-financial social mention frequency analysis and financial social sentiment analysis, which while vastly dissimilar are often mistakenly grouped as one in the same. Financial social sentiment analysis attempts to derive trading signals from a relatively small universe of influential financial chatter and related social sentiment. On the other hand, non-financial social mention frequency analysis involves mining, cleansing, validating, correlating, and analyzing terabytes of unstructured non-financial social data to derive actionable trading signals from interrelated non- financial information determined via mapping to be meaningfully relevant to a specific underlying security, sector, or market.



Applications for non-financial social mention frequency analysis generally fall into three categories:

 

1. Investment Idea Formation / Event Detection 

Surface meaningful fluctuations in conversational frequency providing the earliest detection of off-radar change that could impact a key underlying metric of a business, sector, or macro-economic indicator.   

Arbitrage conventional forms of information dissemination where time frames for broader market digestion, as reflected in the underlying asset price, can range from minutes to weeks.

 

2. Risk Management

Initiate critical and timely hedges against existing equity/credit exposure where, when there is an irregularity in conversational frequency, execution of such a hedge is either performed manually via a human sense check or automatically via an algorithm based on an accepted/validated pre-determined relationship between conversational data and metric/asset price.

As a real time barometer of change, conversational data can also be used to confirm the absence of change, making it a powerful tool for risk management even in quarters where it does not directly lead to hedges. Regular, or conforming conversational frequency would suggest low relative likelihood of unknown or off-radar factors that could adversely impact one's existing investment thesis, supporting one’s trade conviction.

 

3. Research

Derive differentiated intra-quarter insights driven by conversational frequency and sentiment data identified to have high correlations to known key performance indicators or market pricing at the company, sector, or macro level.

Understand when social data and its corresponding narrative are either meaningfully supportive or in contradiction to an existing investment thesis or consensus KPI estimate.

The discussion will examine recent case studies exemplifying a range of successful and flawed social listening approaches for capital markets. Learn how social data sets have been predictive indicators of streaming service subscriber growth, mobile game downloads, corporate adoption of cloud computing, retailer foot traffic, ecommerce sales, box office revenue, and even storm related housing damage.


Chris Camillo, Social Arbitrage Investor, Co-Founder/Advisor at TickerTags by M Science

Chris Camillo

Social Arbitrage Investor, Co-Founder/Advisor
TickerTags by M Science

STREAM A

1:50 pm - 2:05 pm Solving the most typical technical challenges in alt data with machine learning techniques

Gene Ekster, Co-Founder at Alternative Data Group

Gene Ekster

Co-Founder
Alternative Data Group

STREAM A

2:05 pm - 2:20 pm Latest advances in satellite data use-cases
Financial investors, corporations and public sector organizations are exposed to data latency, source biases and market opacity. I will review use cases powered by the GO platform that provide unique data that enables investors to make better informed decisions and address these challenges.
GO is a versatile geospatial analytics platform supporting user-defined criteria for queries that can range from hyper-local to global-scale monitoring. GO empowers users to obtain real-time advantages through analyzing petabytes of raw data covering the global economy, sector by sector, as they occur. 

Users simply define their monitoring criteria and are able to automatically analyze and interpret millions of data points, including: 

• Object Detection
Identify and quantify activities based on data, including car, ship, and aircraft counts sourced from satellite imagery. 

• Geolocation
Access daily foot traffic and key demographic trends, anonymously aggregated from billions of mobile phone location pings. 

• Land Use
Automate large scale change detection of building, road, and environmental land use. 
Bryan Yates, General Manager / Head of Sales, EMEA at Orbital Insight

Bryan Yates

General Manager / Head of Sales, EMEA
Orbital Insight



Lisa Schirf, Former COO Data Strategies Group and AI Research at Citadel

Lisa Schirf

Former COO Data Strategies Group and AI Research
Citadel

Chris Camillo, Social Arbitrage Investor, Co-Founder/Advisor at TickerTags by M Science

Chris Camillo

Social Arbitrage Investor, Co-Founder/Advisor
TickerTags by M Science

Gene Ekster, Co-Founder at Alternative Data Group

Gene Ekster

Co-Founder
Alternative Data Group

Sam Livingstone, Head of Data Science at Jupiter Asset Management

Sam Livingstone

Head of Data Science
Jupiter Asset Management

Bryan Yates, General Manager / Head of Sales, EMEA at Orbital Insight

Bryan Yates

General Manager / Head of Sales, EMEA
Orbital Insight

STREAM A

3:10 pm - 3:30 pm Data labelling and data collection from messy or unstructured sources
This session will discuss how to create high quality labelled data for machine learning and how to build bespoke, multi-dimensional datasets from messy or unstructured source material. Issues covered will include the advantages of building such datasets, the merits of a man/machine approach, how to pick a workforce, and a brief review of a couple of case studies.
Daniel Mitchell, Chief Executive Officer at Hivemind

Daniel Mitchell

Chief Executive Officer
Hivemind

STREAM B

1:30 pm - 1:35 pm QUANT FOR FUNDAMENTAL

Peter Davaney-Graham, VP, Product Strategy, Content and Technology Solutions at FactSet

Peter Davaney-Graham

VP, Product Strategy, Content and Technology Solutions
FactSet

STREAM B

1:35 pm - 1:55 pm Data therapy: The biggest data challenge is the fundamental analyst
Data is not just another tool for fundamental investors, it is a new way of thinking probabilistically about the world and scientifically rejecting hypotheses. This is not a talk about data, it is a talk about humans thinking less instinctively and more purposefully. In other words, can fundamental analysts evolve into data-driven cyborgs? Matei will discuss the challenges and insights gained by working directly with hedge fund analysts for over a decade. 
Matei Zatreanu, Chief Executive Officer at System2

Matei Zatreanu

Chief Executive Officer
System2

STREAM B

1:55 pm - 2:15 pm AI framework for alpha generation and dynamic risk controls

Mark Antonio Awada, Chief Risk & Data Analytics Officer at Alpha Innovations

Mark Antonio Awada

Chief Risk & Data Analytics Officer
Alpha Innovations

STREAM B

2:15 pm - 2:35 pm Improving the investment decision system based on collaborative intelligence
Within the context of (investment) decision making our reliance on human judgment alone, unaided by computers, is going to decrease. Furthermore relying fully on algorithms and data driven insights is also not without challenges. Traditional wisdom says a combination of human and machine should yield better decision outcomes. But which model of collaboration between humans and algorithm is most ideal? What challenges one might face in building such a collaborative model? How can we translate the insights from data science and behavioral science into a robust and repeatable investment process? In this talk I will address some of the above points within the context of multi asset investing.
Prasenjeet Bhattacharya, Lead Data Scientist - Multi Asset at NN Investment Partners

Prasenjeet Bhattacharya

Lead Data Scientist - Multi Asset
NN Investment Partners

STREAM B

2:35 pm - 2:50 pm The automated signal discovery technology empowering traditional analysts
Market participants are observing faster markets and exponential growth in the number of novel datasets. The complexity and the cost of incorporating these datasets are major obstacles for adoption. This is further exacerbated by the ‘talent war’ for top data scientists.
An automated signal discovery technology can empower traditional analysts to search and discover valuable signals for their trading strategy prior to committing to purchasing datasets or investing any resources in data analysis.
The ultimate signal discovery technology is incredibly difficult to build and involved solving previously unsolved scientific challenges such as autonomous data cleaning and causality. The latest advancements in AI may hold the key to unlocking the potential. 
The session will offer some exciting insights and practical use-cases of this powerful technology. 
Darko Matovski, Co-Founder and CEO at CausaLens

Darko Matovski

Co-Founder and CEO
CausaLens


Peter Davaney-Graham, VP, Product Strategy, Content and Technology Solutions at FactSet

Peter Davaney-Graham

VP, Product Strategy, Content and Technology Solutions
FactSet

Charles-Albert Lehalle, Head of Data Analytics at Capital Fund Management

Charles-Albert Lehalle

Head of Data Analytics
Capital Fund Management

Paolo Puggioni, Director, Data and Innovation at BlackRock

Paolo Puggioni

Director, Data and Innovation
BlackRock

Pier Francesco Procacci, Senior Associate, Quant Research at Citi Research

Pier Francesco Procacci

Senior Associate, Quant Research
Citi Research

Aner Ravon, Co-Founder & Chief Strategy Officer at Zirra

Aner Ravon

Co-Founder & Chief Strategy Officer
Zirra

STREAM C

1:30 pm - 1:35 pm AI/ML CUTTING EDGE ACADEMIC RESEARCH

Mehrzad Mahdavi, CEO at Financial Data Professionals (FDP) Institute

Mehrzad Mahdavi

CEO
Financial Data Professionals (FDP) Institute

STREAM C

1:35 pm - 1:55 pm Enhancing time series momentum strategies using deep neural networks
While time series momentum is a well-studied phenomenon in finance, common strategies require the explicit definition of both a trend estimator and a position sizing rule. In this paper, we introduce Deep Momentum Networks -- a hybrid approach which injects deep learning based trading rules into the volatility scaling framework of time series momentum. The model simultaneously learns both trend and position sizing in a data-driven manner, with networks directly trained by optimising the Sharpe ratio of the signal. Backtesting on a portfolio of 88 continuous futures contracts, we demonstrate that the Sharpe-optimised recurrent neural network improved traditional methods by more than two times in the absence of transactions costs, and continued outperforming even when considering transaction costs up to 2-3 basis points. To account for more illiquid assets, we also propose a turnover regularisation term which trains the network to factor in costs at run-time.
Stephen Roberts, Director and Professor of Machine Learning - Oxford-Man Institute of Quantitative Finance, University of Oxford at Co-Founder - Mind Foundry

Stephen Roberts

Director and Professor of Machine Learning - Oxford-Man Institute of Quantitative Finance, University of Oxford
Co-Founder - Mind Foundry

ABN AMRO Clearing Bank works with considerably large amounts of data every day. Using this data we design and deploy in production deep learning models together with hyper-parameter optimization models to approach some of their business cases. Our talk will be focused on how to get insight and meaningful lower-dimensional representations of the data. 

Claudi Ruiz Camps, Senior Machine Learning Specialist at ABN AMRO Clearing Bank

Claudi Ruiz Camps

Senior Machine Learning Specialist
ABN AMRO Clearing Bank

Juan Manuel Acevedo Valle, ML Specialist at ABN AMRO Clearing Bank

Juan Manuel Acevedo Valle

ML Specialist
ABN AMRO Clearing Bank

STREAM C

2:15 pm - 2:35 pm Machine learning to monitor hundreds of algorithms on a trading floor
With the rise of automation, being able to give back control to humans at the best moment with the adequate information is a challenge. I will present how machine learning can detect in real-time potential causes of bad performance of hundreds of trading algorithms, grouping  them in ‘meaningful clusters’ so that a trader can adjust their parameters if needed. It will be the occasion to provide guidelines on how to respond to new industrial needs due to having a lot of AI in production.
Charles-Albert Lehalle, Head of Data Analytics at Capital Fund Management

Charles-Albert Lehalle

Head of Data Analytics
Capital Fund Management

STREAM C

2:35 pm - 2:50 pm A leap of faith? Interpretability of deep learning models for stock selection
Over the last five years, Machine learning has become an intriguing yet interesting topic in quantitative investment. Amongst the numerous families and flavors of algorithms, Deep Learning stands as the one which is as fascinating as it is complex. However, the obsession from researchers for this topic is continuously increasing and results show promising territory for empirical asset pricing and stock selection purposes. Despite the significant promises of neural nets in finance, the pros of those algos seemed clouded and sometimes shadowed by the inherent complexity and the resulting lack of embedded transparency that they exhibit. Naturally, this lack of inherent “interpretability” led to the inflated “Machine Learning is a black box” myth, which turned to be a limit in the trust and then the adoption of those algos in production.

This presentation will review the following points:
•Recap the basic notions of Machine learning and their related specificities for ML in finance
•summarising what are the notions of explainability and interpretability in the field of deep learning models for stock selection.
•Introducing a taxonomy of methods for interpretability
•Equity US stocks universe use case to explain the different techniques and results for:
  • global models
  • Local models
  • Partial dependence
•Expanding the use case by focussing on two deep learning models for stock selection
  • Multi-Layer Perceptron (MLP)
  • Convolutional Neural nets (CNN)
Tony Guida, Executive Director - Senior Quant Research at RAM Active Investments

Tony Guida

Executive Director - Senior Quant Research
RAM Active Investments

STREAM C

2:50 pm - 3:10 pm Forecasting the cross-section of stock returns using machine learning algorithms
Forecasting stock returns at the firm level brings the challenge of evaluating the independent information in the entirety of many cross-sectional predictor variables, their potential interactions and non-linearities. While traditional portfolio sorts and simple linear regressions are not up to that task, machine learning algorithms are well suited for that problem. This talk gives answers on when, why and how to use ML algorithms in forecasting stock returns at the firm level.
Benjamin Moritz, Executive Partner at HQ Asset Management

Benjamin Moritz

Executive Partner
HQ Asset Management

STREAM C

3:10 pm - 3:30 pm Quantum computing in finance: Where to apply it

Samuel Mugel, Co-Founder at Multiverse Computing

Samuel Mugel

Co-Founder
Multiverse Computing

3:30 pm - 3:55 pm Networking refreshment break in the exhibition area

STREAM A

3:55 pm - 4:00 pm NATURAL LANGUAGE PROCESSING
Anthony Tassone, CEO at GreenKey Technologies

Anthony Tassone

CEO
GreenKey Technologies

STREAM A

4:00 pm - 4:20 pm Teaching machines to understand Chinese cyberslang

Mike Chen, Director Dynamic Equity at PanAgora AM

Mike Chen

Director Dynamic Equity
PanAgora AM

STREAM A

4:20 pm - 4:40 pm Unlocking internal data using NLP
We generate too many emails, store too many files and have more information than we know about or are able to handle. While conversations across email and messengers are growing senseless, our data is mostly unused, fragmented and decentralized making it impossible to keep track, let alone know exactly what we have. While organizations believe there is value in their own data, the challenge is finding ways to monetize it. In this presentation, Peter Hafez will present a real case study showcasing how a fundamental asset management firm is leveraging RavenPack technology to transform their own emails, messages, and files into trading signals to create alpha generating strategies.
Peter Hafez, Chief Data Scientist at RavenPack

Peter Hafez

Chief Data Scientist
RavenPack



Anthony Tassone, CEO at GreenKey Technologies

Anthony Tassone

CEO
GreenKey Technologies

Mike Chen, Director Dynamic Equity at PanAgora AM

Mike Chen

Director Dynamic Equity
PanAgora AM

Peter Hafez, Chief Data Scientist at RavenPack

Peter Hafez

Chief Data Scientist
RavenPack

Bennett Saltzman, Senior NLP Data Scientist and Lead Model Developer at Amenity Analytics

Bennett Saltzman

Senior NLP Data Scientist and Lead Model Developer
Amenity Analytics

STREAM B

3:55 pm - 4:00 pm MARKET MICROSTRUCTURE & HFT
Jakob Aungiers, Head of Systematic Market Making at Quantstellation Capital

Jakob Aungiers

Head of Systematic Market Making
Quantstellation Capital

STREAM B

4:00 pm - 4:20 pm Use of deep learning for modelling high-frequency market microstructure data
After a brief introduction to market microstructure and high-frequency financial data, we present recent work on a large-scale deep learning model to predict price movements from limit order book (LOB) data of cash equities. The model, which utilises convolutional filters and LSTMs, is trained using full-resolution market data from the London Stock Exchange. Our model delivers a remarkably stable out-of-sample prediction accuracy for a variety of instruments and outperforms existing methods. Interestingly, our model translates well to instruments which were not part of the training set, indicating the model's ability to extract universal features.
Stefan Zohren, Academic Faculty Member at Oxford-Man Institute of Quantitative Finance and University of Oxford

Stefan Zohren

Academic Faculty Member
Oxford-Man Institute of Quantitative Finance and University of Oxford

STREAM B

4:20 pm - 4:40 pm Enabling value extraction from limit order book data
Session preview:
L3 limit order book (LOB) data contains 30% more price discovery relevant information than L1 data. The problem is this data is difficult to work with and the information is hard to extract. While a small handful of very well resourced leading market participants have capitalized on being able to use this data over the last decade, most major financial institutions are still not able to. 
 
Some believe that L3 information is not relevant to those trading through brokers, nor relevant to those trading at medium to low frequencies. To which we would make three observations. Firstly, while it is true that L3 data does hold information on short-term patterns, it is only by down-sampling this data that lower frequency data is obtained. Pre-processing with a zero-one filter is a crude tool, throwing away information whether it is relevant or not. The statistically correct approach is to allow the model to decide how to handle the data. Secondly, the “leading market participants” referred to above generally ply their trade through what many consider to be ‘predatory’ trading. This can mean providing liquidity when it suits, but as soon as a natural order is detected in the L3 data, leaning against that order and taking liquidity – the end result being a cost to the provider of the natural order. Only by understanding the statistical dynamics of parent-child submission through using L3 data can such predatory approaches be defeated. Thirdly is the ability to accurately simulate the market using agent based models (ABMs), trained using L3 data. From trained ABMs synthetic market data can be generated. The ability to simulate medium-low frequency trading strategies not just against a one-time realization of market data, but across many realizations enables a step change in statistical significance of predictor design. 
 
The workflow of using L3 data is as follows: Data is collected by performing packet capture at the colo with potentially every pcap location recording data from every matching engine. This captured data is then centralized and subject to a process of curation (eg book building, ticker mapping), normalization (eg transform to UTC, map fields to API dictionaries) and consolidation (eg enable my view of the European Consolidated Tape). An additional high-value step is combing the L3 public data with private order flow data to generate L4 data – enabling identification of beneficiaries orders in the book, along with other information such as order types and max show values. This data pipeline will include metadata management and potentially derived data management (eg generate intraday volume curves). Once the data is present it then needs to be combined with cheap compute at scale. This is either by an in-house farm or using the cloud, using either map-reduce or non-map-reduce systems. The workflow needs to allow for the range of L3 data use cases and where and how to implement fine grained and coarse grained parallelism to ensure sufficient speed. Finally, the workflow needs access to both open source analytics (eg pandas, tensorflow) and closed source analytics (eg MOSEK). The end user needs to be able to make any arbitrary calculation on any amount of L3 data (eg Russell 3000 for the last three years) and have results returned to them in an appropriate amount of time. At BMLL we see our value as being at the interfaces of this value stack, with the interfaces being represented as APIs. 
 
A recent industry survey of quants found they felt they spent over 80% of their time performing menial duties around data and systems, as opposed to performing their value-add. Another recent survey of employers felt that they were severely under-resourced in the quant and engineering fields. In summary, managed services are powerful as they help solve for both these pain points. Managed services enable human resources to focus on their value add, while also speeding up the OODA loop of complex problem solving.
Hugh Christensen, Founder, Head of Research at BMLL Technologies

Hugh Christensen

Founder, Head of Research
BMLL Technologies

STREAM B

4:40 pm - 5:00 pm Statistical methods to detect market asymmetries
Order book and market event data is a predominant example of a complex system with heterogenous users. We present a very recent work applying statistical physics methods to market microstructure and high frequency trading. We apply methods and approaches from statistical physics to build dynamic null models of market event flow in time. These well-defined distributions are successfully applied to predict flash crashes with minor false positive rates and timely-leading predictive power. We also explore applications to active high frequency trading and build profitable strategies based on market anomalies and imbalances from our null model.
Jeremy Turiel, PhD Candidate - Financial Computing and Analytics Group, UCL at AVP, Barclays Investment Bank

Jeremy Turiel

PhD Candidate - Financial Computing and Analytics Group, UCL
AVP, Barclays Investment Bank

STREAM B

5:00 pm - 5:20 pm Predictive models for HFT

Andrew Mann, Quantitative Researcher - Market Maker at Virtu Financial

Andrew Mann

Quantitative Researcher - Market Maker
Virtu Financial

STREAM C

3:55 pm - 4:00 pm HUMAN CAPITAL
Benoit Mondoloni, Head of PB Analytics at Bank of America Merrill Lynch

Benoit Mondoloni

Head of PB Analytics
Bank of America Merrill Lynch

STREAM C

4:00 pm - 4:20 pm Diverse teams beat homogenous teams every time in terms of performance
Data Science is a team sport. And data science is inherently multi-disciplinary, combining many different skills and approaches. Having built the Data Insights Unit in Schroders over the last 5 years, I will be talking about the many elements of getting a diverse team to work together effectively, including motivating them, organising them, and rewarding them. This will include a few specific techniques, frameworks and tools that we’ve found have really helped, and the lessons we’ve learned along the way.
Mark Ainsworth, Head of Data Insights at Schroders

Mark Ainsworth

Head of Data Insights
Schroders

 How have firms been addressing the talent shortage?
Oliver Blaydon, Head of Advances Analytics and Risk at Armstrong International

Oliver Blaydon

Head of Advances Analytics and Risk
Armstrong International

Mark Ainsworth, Head of Data Insights at Schroders

Mark Ainsworth

Head of Data Insights
Schroders

How can you ensure your team remains up-to-date with the latest techniques and technology?
Benoit Mondoloni, Head of PB Analytics at Bank of America Merrill Lynch

Benoit Mondoloni

Head of PB Analytics
Bank of America Merrill Lynch

Mark Ainsworth, Head of Data Insights at Schroders

Mark Ainsworth

Head of Data Insights
Schroders

Alex Remorov, Vice President, Systematic Active Equities at BlackRock

Alex Remorov

Vice President, Systematic Active Equities
BlackRock

Sue Lai, Head of Product at DataCamp

Sue Lai

Head of Product
DataCamp

Anthony Tattersall, Senior Director of Enterprise, EMEA at Coursera

Anthony Tattersall

Senior Director of Enterprise, EMEA
Coursera

5:20 pm - 5:30 pm Closing remarks

5:30 pm - 7:00 pm Drinks reception and networking in the exhibition area