Machine Learning News & Topics for Quantitative Trading and Algorithmic Development

Machine Learning (ML) is the evolution of artificial intelligence where the computer (program) works with data to discover patterns (also called features) that can be used later to evaluate other data.  ML is typically broken down into three categories: supervised ML, unsupervised ML, and reinforcement learning.

CloudQuant does enable Machine Learning. To learn more check out “An Intro to Machine Learning with CloudQuant and Jupyter Notebooks” by Trevor Trinkino. This post and video cover one quantitative trader’s approach to supervised ML.

Posts

Morgan Slade, Python Data Scientist and Trader

Quant Trading and Superpowers: Morgan Slade speaks on Opportunity

,
“You have a chance to try and change an industry” said Slade, CEO of CloudQuant at the MarketsWiki Education’s World of Opportunity event in New York.
Quantitative Strategies and Capital for Trading

Quantitative Trading and Data Science in the News August 14 2017

Topics include: GeoLocation Alternative Data, robotic revolution, buy side, sell side, hot jobs, financial crime, …
Quantitative Strategies and Capital for Trading

Quantitative Trading and Data Science in the News August 7 2017

August 7, 2017

Citadel’s Flagship Funds Gain Almost 7% This Year

…Citadel’s Tactical Trading Fund, which uses equity and quantitative strategies, rose 3 percent last month, bringing year-to-date performance through July to 4.9 percent. …  

Hedge funds lose more than half a billion on wrong-way bet against Tesla

It’s not just hedge funds that bet the wrong way on Tesla. Wall Street analysts, normally a very bullish crowd, were largely negative on the stock heading into the earnings report. They reiterated their bearishness in reports on Thursday, despite the stock pop. “We were surprised by the after hours move in TSLA shares and continue to be cautious on the stock, especially as the risk profile shifts from the hype of the Model 3 to execution, or ‘production hell’ as Elon Musk refers to it,” Cowen analyst Jeffrey Osborne wrote in a note. CloudQuant note — We wonder what the Algos were choosing? Were they different than the analysts?  

Watson Machine Learning is now Generally Available

IBM announced the general availability of the IBM Watson Machine Learning service. Over the past 12 months feedback from hundreds of users of the Watson Machine Learning (WML) service led to this announcement. CloudQuant note — We love seeing more people able to advance the cause of Data Science and Machine Learning  

Google chief funds new machine-learning effort at Princeton’s IAS

A $2 million donation will launch new research at the Institute for Advanced Study (IAS) in Princeton to forge an understanding of how machine learning evolves. Machine learning — sometimes called the leading edge of artificial intelligence — is the rapidly developing computer technology behind self-driving cars, complex web searches, medical and science applications, and face and speech recognition. Machine-learning programs synthesize knowledge in a way that’s analogous to how children learn. The programs take examples, generalize, and then develop rules and understanding about the world without being taught directly. With time, the programs become better at particular tasks. CloudQuant note — We love seeing academic chances to advance the cause of Data Science and Machine Learning  

10 hot data analytics trends — and 5 going cold

Big data, machine learning, data science — the data analytics revolution is evolving rapidly. Keep your BA/BI pros and data scientists ahead of the curve with the latest technologies and strategies for data analysis.
CloudQuant note — We are definitely in agreement on the topics of Scikit-learn, TensorFlow, and Jupyter Notebooks
SPY Benchmark Trader

Machines Poised to Take Over 30% of Work at Banks, McKinsey Says" on Bloomberg

New technologies are poised to sweep through investment banks, relieving many rank-and-file employees of roughly a third of their current workload, according to McKinsey & Co. The shift, already stoking angst on Wall Street, may take only a few years. Cognitive technologies — applications or machines that perform tasks once requiring human thought — are now cheap enough that banks can deploy them across operations facilitating trades or other capital-markets business. In a report Thursday, McKinsey said automating tasks will “free up capacity” for staff to focus on higher-value work, such as research, generating new ideas or tending to clients. “This is really starting to take steam and it’s going to transform the industry over the next two to three years,” Jared Moon, a McKinsey partner who co-wrote the report, said in an interview. The consultants estimate cognitive technologies will free 20 to 30 percent of employees’ capacity in units processing trades.
Automation has been sending shivers down spines across Wall Street, as workers worry they will be replaced by machines …. Read the full article “Machines Poised to Take Over 30% of Work at Banks, McKinsey Says” on Bloomberg – July 20, 2017, 7:00 AM CDT  

CloudQuant Thoughts

We see opportunities in this change. The growth of automation in the trading and banking industry is creating opportunities for knowledgeable workers to transition from old jobs. This is similar to the move that allowed many traders to transition from the trading pits to electronic trading. Those that choose to innovate and re-create their careers are able to move into quantitative investing. All they need is a willingness to learn, to work hard, and to have access to capital.
World Market Access

2017 – The Year of Artificial Intelligence

2017 is the year of artificial intelligence. Here’s why

World Economic Forum published that Artificial Intelligence (AI) is a rapidly growing discussion point in corporations and governments. This is driven by: 1. Everything is now becoming a connected device

The internet of things is collecting data in ways never before possible.

2. Computing is becoming free

The cost of computing continues to drop, especially with crowdsourced research platforms like CloudQuant.

3. Data is becoming the new oil

“The amounts and types of data available digitally have proliferated exponentially over the last decade, as everything has moved online, been made mobile with smartphones, and tracked via sensors. New sources of data emerged through things like social media, digital images and video.” 

4. Machine learning is becoming the new combustion engine

“new machine learning models have emerged recently that seem to be able to take better advantage of all the new data. For example, deep learning enables computers to ‘see’ or distinguish objects and text in images and videos much better than before.”

At CloudQuant our crowd researchers are finding that access to markets, and to data is allowing them to research and develop profitable algos in ways never before conceived. Access to new data sets, like social sentiment, allow new dimensions of quantitative strategies that were not conceived even five years ago. We anticipate that the new data “oil” and machine learning “engines” will continue to grow our world of trading.   See the full article on World Economic Forum’s web site by Sandhya Venkatachalam (24 May 2017).
Battle of the Quants June 2017

Battle of The Quants – Discusses Crowd Researching in NY

,
Crowdsourcing in fund management and trading is the move to utilize anyone with an internet connection to participate in the research with the goal of finding new and better ways of trading. During the discussion the differing approaches being taken with the business models, and the technology, and the challenges each are facing.
Python based Trading Strategies by Machine Learning

Trading Strategy development—Powered by Machine Learning

,
Morgan Slade, the CEO of CloudQuant spent some time with discussing the world of crowd research, technology, and data science with ChatWithTraders.com Topics covered:
  • How large funds and institutions put on $100-million positions; how they work orders into the market, structure the trade and handle market impact etc.
  • Morgan explains why he feels as though the common approach to strategy development is counter intuitive, and shares an alternative 3-step formula.
  • A simple description of how machine learning and data science is being used by traders, and an example of how ML has been used to improve existing strategies.
This interview is the follow on interview that Chat With Traders had Andy Kershner. Towards the end of that episode, Andy briefly mentioned a cloud-based algo development platform and fund, CloudQuant, which is a subsidiary of Kershner Trading Group.

About Morgan Slade

Learn more about Morgan, his 20 years of experience as a trader, portfolio manager, researcher, technologist, executive and entrepreneur, and the team at CloudQuant.
Quantitative Strategy, Trading, and Algo Development Industry News

An Index-Fund Evangelist Is Straying From His Gospel

In his classic 1973 book “A Random Walk Down Wall Street,” Burton Malkiel, a Princeton economics professor, made an assertion that was startling at the time: that “a blindfolded monkey throwing darts at the stock listings could select a portfolio that would do just as well as one selected by the experts.”

Three years later, Vanguard, the asset manager where Mr. Malkiel served on the board for 27 years, started the first passive index fund, an innovation that has swept the financial world.

Now, at age 84, Mr. Malkiel has had a remarkable change of heart: Maybe the experts can beat the monkeys after all. That is, if the experts are software engineers writing sophisticated algorithms for computer-generated trading.

A large and growing body of academic research suggests there are market anomalies that can be exploited to beat a strict index approach. Some of that research has been recognized with Nobels in economic science — William F. Sharpe in 1990 and Eugene F. Fama in 2013. One of these findings is that value outperforms growth, rewarding those who identify stocks with lower price-earnings ratios and other metrics that suggest they’re undervalued. Another factor is momentum, in which stocks that are already outperforming market averages continue to do so.

There’s a lot of statistical and, perhaps more important, behavioral support for these strategies,” Mr. Hougan said. “You’ll find plenty of two- or three-year stretches where this will underperform, but if you buy and hold, it’s going to add value. We’ve seen value outperform for over 80 years. And Wealthfront is blending five factors that should smooth out and reduce those periods of underperformance.”

Read the full article on the New York Times

Stock Market, Quantitative Strategy, Trading, and Algo Development Industry News

Social Sentiment in Trading Algorithms

,
Bloomberg recently wrote that “It’s no secret that hedge fund managers are always looking for new sources of data that will help them in their never-ending quest to beat the market.” (1) One of the most interesting new sources of data is social sentiment. We have found that the incorporation of social sentiment data is definitely improving the quality of algorithms as shown in our backtesting on CloudQuant. Over the next couple of weeks an intern from the University of Chicago who is mastering in Financial Mathematics is working on incorporating social signals into a DeMark Indicators script that is available for all registered users to see in the CloudQuant base working scripts. I look forward to seeing how this improves. And I look forward to seeing her quantitative reasons for why social sentiment and other changes to the TD Sequential script improves. (1) Finding Novel Ways to Trade on Sentiment Data | Tech At Bloomberg
Stock Market, Quantitative Strategy, Trading, and Algo Development Industry News

Machine learning set to shake up equity hedge funds – Financial Times

AI seen becoming powerful enough to forecast market moves better than humans

Financial Times May 25, 2017 by: Lindsay Fortado and Robin Wigglesworth
Machine learning poses a threat to equity hedge funds within the next decade as the technique becomes powerful enough to forecast market moves better than humans, one of the earliest investors in the industry is forecasting. Jeff Tarrant, the founder of Protégé Partners, says that the model of hedge funds charging “2 and 20” — a 2 per cent management fee and 20 per cent performance fee — for investing in large-cap stocks rising and falling “doesn’t work any more” and is ripe for disruption. He pointed to the overhaul of other industries in the past decade at the hands of engineers and scientists. “Jeff Bezos picked off the bookstore business. Apple totally picked off the music business and Netflix totally changed television. Now [machine learning] is going to pick off the hedge funds.”
Read the full story on Financial Times