Javascript Menu by Deluxe-Menu.com
World Business Strategies Logo

09.00 - 09.10: Welcome Address. Chairman’s Opening Remarks

08:30: Morning Welcome Coffee

Chair:Image result for yves hilpisch

Yves Hilpisch

Founder and Managing Partner

The Python Quants


09.00 - 10.00: Keynote SpeechMarcos Lopez de Prado

Presenter: Marcos Lopez de Prado: Research FellowLawrence Berkeley National Laboratory

The 7 Reasons Most Machine Learning Funds Fail

Over the past 20 years, I have seen many new faces arrive to the financial industry, only to leave shortly after. The rate of failure is particularly high in machine learning (ML). In my experience, the reasons boil down to 7 common errors:

1. The Sisyphus paradigm
2. Integer differentiation
3. Inefficient sampling
4. Wrong labeling
5. Weighting of non-IID samples
6. Cross-validation leakage
7. Backtest overfitting


10.00 – 10.45: Artificial Intelligence and the Future of Capital Markets Investing

  • Why Artificial intelligence for Capital Markets investing?
  • When Artificial Intelligence proliferates; won’t returns be arbitraged away?
    • Challenge #1 Data Acquisition, Integration, Processing Power
    • Challenge #2: Artificial Intelligence and its subcomponents 
  • Where should financial services professionals focus their effort?

Presenter: Michael Beal: CEO, Data Capital Management


10.45 – 11.15: Morning Break and Networking Opportunities


11.15 - 12.00: Machine Learning for Trading Gordon Ritter

Abstract: In multi-period trading with realistic market impact, determining the dynamic trading strategy that optimizes expected utility of final wealth is a hard problem. In this paper we show that, with an appropriate choice of the reward function, reinforcement learning techniques (specifically, Q-learning) can successfully handle the risk-averse case. We provide a proof of concept in the form of a simulated market which permits a statistical arbitrage even with trading costs. The Q-learning agent finds and exploits this arbitrage.

Presenter: Gordon Ritter: Senior Portfolio Manager, GSA Capital  


12.00 - 12.45Statistical Algorithm Selection: A Data Science Approach to Managing Systematic Trading Strategies Developed by "The Crowd"Jess Stauth

Abstract: 

The field of quantitative finance has experienced rapid adoption of machine learning techniques at nearly every stage of the typical workflow, e.g. new signal identification, signal combination, portfolio optimization and trading. 

We explore how the techniques of machine learning might be applied to a unique new problem: identifying trading strategies which are likely to produce "alpha" in a real market setting from a pool of millions of backtests created by Quantopian's 170,000+ member online community. We will review challenges including:

- compiling our (imbalanced) data set of simulation results

- designing features based on transactions, returns, holdings, and more?

- defining a goal (what should we select for?) 

- facing the problems of non-stationarity, overfitting, and interaction terms 

Presenter: Jess Stauth: Managing Director, Quantopian  


12.45- 13.45: Lunch


13.45 - 14.30: AI-First Finance and Algorithmic Trading

Image result for yves hilpisch

  • This talk considers the consequences of recent advances in the field of Artificial Intelligence (AI) for finance in general and algorithmic trading in particular.

  • The talk is mainly based on practical examples, using Python as well as Machine & Deep Learning techniques to come up with algorithmic trading strategies.

  • The examples in turn are mainly based on data from the recently released Thomson Reuters Eikon Data API for data retrieval and financial analysis. 

Presenter: Yves Hilpisch: Founder and Managing Partner, The Python Quants


14.30 – 15.15: Machine Learning & Event Detection for Trading Energy Futures 
Image result for peter hafez ravenpack

Abstract: The emergence of big data in finance has had a major impact on equities trading. However, other asset classes have seen less of an impact, since fewer alternative data sets are available to support these. In recent years this has changed with the proliferation of various social media sources and with the development of more advanced knowledge graphs that support a global macro theme. During this talk Peter Hafez will show how RavenPack Analytics (RPA) can be used to uncover profitable trading signals for commodities based on news events detected across thousands of sources. In particular, he utilizes ten well-known machine learning algorithms to predict next day returns across a broad energy commodity basket. 

Presenter: Peter Hafez: Chief Data Scientist, RavenPack 


15.15 - 15.30: Afternoon Break


15.30 - 16.15: Extracting Embedded Alpha in Stocks and Commodity Underlyings using Statistical Arbitrage/ML Techniques from News/Social Data

Arun Verma

  • Extracting actionable information in the high volume, time-sensitive environment of news and social media stories
  • Using machine learning to address the unstructured nature of textual information
  • Techniques for identifying relevant news stories and tweets for individual stock tickers and assigning them sentiment scores
  • Demonstrating that using sentiment scores in your trading strategy ultimately helps in achieving higher risk-adjusted returns
  • Illustrate results for both stocks and commodities.

Presenter: Arun Verma: Quantitative Research Solutions, Bloomberg, LP