Sentiment Analysis Week 16 – 20 June

UNICOM is pleased to announce their annual Behavioral Models and Sentiment Analysis Applied to Finance Conference and Workshop Series
Sentiment Analysis Conference, London, Millennium Mayfair, 18 – 19 June 2014: This conference is a one of a kind event ; it looks specifically at how media sentiment (news wires , social media micro blogs, twitter) impact the financial markets. UNICOM has brought together a unique group of academic and industry researchers whose work is exclusively focused in this domain. This one-week event series would be of great value to quant research teams within Hedge Funds, Investment Management and Trading desks of investment banks.

The major finance industry players such as Bloomberg, Thomson Reuters, RavenPack, Northfield Information Services, Deltix and Market Psych are participating and presenting their analytic research and services products.

In addition to presentations by leading academics from USA, UK , Europe and Asia there are three panel sessions on the following topics:

1. Adoption of Sentiment Analysis for Fund Management and Trading Strategies
2. Use of News and Social Media Data for Market Surveillance & Operational Risk Control
3. Data Sources: Market Data, News (Meta) Data, Social Media Data, Micro blogs(Twitter), Google Trends & others

For full details of the programme please visit http://conferences.unicom.co.uk/sentiment-analysis/programme.php

Pre and Post Conference Workshops, London, Fitch Learning, 16 – 17 & 20 June 2014: These workshops are hosted by industry professionals and cover the topics of:
1. Market Microstructure, Liquidity and Automated Trading – 16 & 17 June
2. Sentiment Classification and Opinion Mining Using News Wires and Micro Blogs – 20 June

Posted in Uncategorized | Tagged , , , | Leave a comment

Sentiment Analysis Applied to Finance

Sentiment Analysis Applied to Finance

Sentiment Analysis has developed as a technology that applies machine learning and makes a rapid assessment of the sentiments expressed in news releases. News (events) move the market and are measured quantitatively. Analysts and investors digest financial news and their perceptions impact the market and move stock prices. This conference presents the current state of the art in this fast-emerging field. The conference also presents the current state of knowledge in the application of Sentiment Analysis to the respective models of trading, fund management and risk control.  Major news (meta) data suppliers such as Bloomberg, Thomson Reuters, Dow Jones, RavenPack and MarketPsych have committed their participation.  . Case studies by investment banks, proprietary trading houses and financial analytics providers are under discussion; further such contributions are solicited.  Leading academics, thought leaders and researchers from Europe, UK and USA have agreed to contribute and participate in the conference and the workshops.

A series of webinars are planned to introduce the key topics:

Provisional Time Table for Webinars for Sentiment Analysis Week 16 – 20 June, 2014

Date Presenters
16  April  Stephen Pulman, Oxford Unv (Confirmed)
29 April

 

RavenPack (Confirmed)
7 May Dan diBartolomeo, Northfield Info Serv (confirmed)
14 May Jacob Sisk, Thomson Reuters (TBC)
21 May Bloomberg (TBC)

The weeklong event (16 – 20 June) is specially commissioned for the Fin-Tech professionals for the BFSI sector and will consist of two workshops and a two day conference

In recent times, the use of social media, google trends in addition to automated machine readable news has radically altered the landscape of automated trading, trade surveillance, fund management and risk control for the major market participants.

This event brings together from the industry and academia the best of breed contributions who present the latest thinking covering these cutting edge topics.

 

Posted in Uncategorized | Leave a comment

OptiRisk sponsored Big Data Conference 2013 at UNICOM Seminars

Just giving a quick heads up of OptiRisk’s latest activities. OptiRisk Systems sponsored and exhibited at “Big Data in the Context of Financial Analytics & Social Media: B2B & B2C Infonomics” on 5 December.

Let me give some background to Big Data. Nowadays the Big Data word is buzzing around in nearly all organisations, and there is a pressing need to implement big data strategies. Data in the cloud provides the opportunity to further explore Big Data analytics and helps them to decide which data they should save forever and which they should throw away. 

Big Data in the Context of Financial Analytics & Social Media: B2B & B2C Infonomics” conference was organised by UNICOM Seminars on 5 December at Novotel London West. Many renowned speakers presented their experience of Big Data in the conference.

The conference was welcomed and introduced by Prof. Gautam Mitra, CEO of OptiRisk Systems. Professor David J Hand, Chief Scientific Advisor at Winton Capital Management and an emeritus Professor of Mathematics at Imperial College London, was the opening keynote speaker, who presented “Big Data: less than meets the eye?” He spoke of the potential benefits of Big Data, Open Data as well as the rules. The audience liked the summary of his talk that  “no one actually wants data; they want the answers which can be extracted from data”.

David’s talk was followed by Mike Ferguson, Intelligent Business Strategies, who gave a presentation on “Big Data: Multiplatform Analytics” and explained the critical factors of big data. Dai Clegg from Acunu presented “Rise of Real Time Analytics: the second wave of Big Data”. Other presenters were Daniel Roberts from MongoDB, Charlie Hull from Flax, Cynan Rhodes & Clive Shirley from FSWire and the last presenter Federico Alberto Pozzi joined the conference from University of Milano-Bicocca in Italy.

This conference was organised alongside “Adopting Open Source Software within the Corporate ICT Strategy”. There was a combined panel session: Open Source Software & Proprietary Software: Identifying the winning strategy” across the both conferences. This panel was moderated and chaired by Prof. Mitra and panellists included Daniel Roberts, MongoDB; Clive Shirley, FSWire; Charlie Hull, Flax; Matt Hamailton, Netsight: Hugo Trippaers, Schuberg Philis.

OptiRisk Systems thanked UNICOM for organising this event and for bringing together leading finance professionals in the sector of big data. We would welcome your feedback of this blog.

Posted in Workshops and Seminars | Tagged , , , , , , | Leave a comment

Adopting Open Source Software within the Corporate ICT Strategy

Open Source Software has facilitated innovation in the market place. It is already a backbone of cloud innovation and has hastened the delivery of social solutions in Customer Relationship Management (CRM) and provided applications for development in many industries.

Most Open Source projects are hosted by non-profit organisations. Many of them have the word “foundation” in their name – Mozilla Foundation, Document Foundation, Apache Software Foundation and collectively they are known as “foundations”. Business leaders use to discuss matters of mutual interest – called Floss (Free/Liber and Open Source Software). The important aspect of all Floss licensing is to grant a license that gives the same rights and freedoms to everyone in the public – individuals, non-profit and profit organisations. Government agencies are increasingly turning to Open Source technologies in the wake of shrinking budgets.

Enterprises are dealing with exploding levels of business events and associated data. In order to stay competitive and meet customer expectations, the companies need to embrace new standard for communication on the internet and Open Source has made it easy for them. There is licensing strategy for Open Source Software. Open Source Software (OSS) licensing permits businesses to open up their projects to the public access and they can submit, add, change, repair, or to build on top of the existing software code. Many organizations or developers do not have a positive understanding of the differences between the various OSS licenses and how to go about selecting one that fits closely with their software project, nor are they well-informed of how they want the project to be open for public involvement.

UNICOM is organising a conference “Adopting Open Source Software within the corporate ICT strategy” on 5 December. This conference sets out to attract the best of breed open source proponents and those government departments, NGOs and multinationals who see the growing market strength of the OSS movement and plan to build this into their formal ICT strategy.

Posted in Uncategorized | Leave a comment

Sentiment Analysis Applied to Finance Conference

Last week UNICOM, in association with OptiRisk Systems organised a conference “Behavioural Models and Sentiment Analysis Applied to Finance” (2 – 3 July). Over 85 professionals from the finance industry attended the event which was hosted by FitchLearning one of the sponsors of the event. In addition 20 other professionals from all over the world attended the conference virtually, that is online. Altogether the participants included, CEOs, Senior Quants, Senior Portfolio Managers, Global Heads of Quantitative and Derivatives, Traders and many more from Asset Management, Investment Management firms, trading, banking and insurance sector. The conference was sponsored by RavenPack, the leading provider of real-time news analytics services.  At the conference itself, the blend of the industry & academic sector explained how to create a clear strategy for sentiment analysis applied to finance; the theory and case studies provided exemplars for the finance sector to this new approach and extend its reach.  

The first day of the conference was opened by Prof. Gautam Mitra and chaired by Dan diBartolomeo. The day began with a presentation “Text Mining for Sentiment Analysis” by Sanjiv Das, Professor of Finance, Santa Clara University, followed by Peter Hafez, Director of Quantitative Research, RavenPack, who highlighted “Market – Level Sentiment for Trading Forex and Equity Indexes” in detail.  The contents of the conference were striking and appealed to the audience. The fascinating and inspiring panel sessions engaged people actively in the discussion. The first panel was chaired by Armando Gonzalez, CEO, RavenPack; other panellists include Greg B Davies, Sanjiv Das, Gautam Mitra and Gurvinder Brar. The panel discussed “Adoption of Sentiment Analysis in Fund Management and Trading Strategies”. Armando highlighted that over the last decade, financial firms have begun embracing alternative sources of information to make investment and trading decisions. In particular, machine-readable news and analytics have made their way into quantitative and algorithmic trading programs.

Greg B Davies, Managing Director / Head of Behavioural Investment Philosophy, Barclays Wealth gave a presentation on “Maximising Anxiety Adjusted Returns” and Sanjiv Das in his second talk presented “Network Finance Models”. Gurvinder Brar, Head of Quantitative Research Group, Macquarie explained “Text Mining for Longer – Horizon Investors” and Gautam Mitra, CEO  & Founder, OptiRisk Systems highlighted “Impact of News Events and its Application in Trading Strategies”. The second panel of the conference “Use of Sentiement Data for Risk Control and Compliance” was chaired by Peter van Kleef, CEO, Lakeview and other panelists include Dan diBartolomeo, Mats Wilhelmsson, Chrisol Correia. Panel 2 was followed by drinks and networking. The event delivered a unique networking opportunity for leveraging finance and structured finance professionals to meet in one location and explore opportunities in interconnected finance markets.

The second day of the conference, 3 July, was chaired by Gautam Mitra and Richard Peterson, CEO, MarketPsych fascinated the delegates with his talk “Inside the Global Brain: How sentiment trends in news and social media influence global equity and currency values”. The second talk “How to use Language Recognition and Quantifiable news flow in creating trading strategies” was delivered by Marco Dion, Managing Director, Global Head of Equity Quant Strategy, J P Morgan. Then Dan diBartolomeo , President & Founder, Northfield Information Services Inc, gave a presentation on “Credit Risk Assessment of Corporate and Bank Debt using Sentiment and News”. In the conference two academic researchers, Xiang Yu from Brunel University and Giles-Arnaud Nzouankeu Nana from Fraunhofer ITWM presented their research work. Xiang presented “Introducing Sentiment in the Predicitive Analysis of Asset Behaviour: Return, Volatility and Liquidity in an Intraday Setting” and Giles presented   “Improvement of the Volatility Measurement through inclusion of Market Behviour”.

One of the interesting parts of the conference was the presentation of Rochester Cahan, CFA, Director / Head of US Quantitative Strategy, Deustche Bank Securities Inc, who explained “Quant 3.0: Harnessing the mood of the web in alpha strategies”. Rochester presented to the conference live from New York.  The last panel of the conference was chaired by Richard Peterson and other panelists include Rochester Cahan (online), Peter Hafez, James Cantarella and Bram Stalknecht. This panel gave very interactive discussion on “Social Media, Metadata, Trust, Information Leakage”. Both days of the conference ended with a very good panel question and answer sessions. The panellists and presenters made some significant commitments for sentiment analysis strategies and these could have a very positive impact in the long run.

The feedback from the participants was extremely positive and participants remarked that the event was creative and resourceful. Some commented that it was one of the best conferences on Sentiment Analysis Applied to Finance to be held in the UK. All doors of discussion in the conference were opened and every idea presented in the conference was welcomed and appreciated. The event provided a platform to the sell side companies of providing news and sentiment analysis data and also to the buy side companies to discuss with data providers directly. The conference had a fantastic buzz throughout the two days. The speakers and panellsits throughout were of high calibre, as many delegates commenting that the day was inspiring and motivating for them.

Thank you to everyone who attended the conference and also a big thank to our sponsors RavenPack and CQF. Thank to UNICOM to organising a first ever conference on “Behavioural Models and Sentiment Analysis Applied to Finance” to bring together leading finance professionals and those who want to go ahead in this market. I would welcome your feedback of this blog. The conference material and the full recording of the conference are available online. Please contact info@unicom.co.uk.

Author:

Aqeela has completed her MSc. in Risk Management and Modelling from Brunel University.  She has previously worked as a lecturer in Physics, as a credit Officer at MCB Bank and as an administrator for FIFA. She has broad experience in quantitative modeling, market and credit risk management and analysis. She is actively involved in lead generation, identification of business growth areas, brainstorming on new business opportunities and following up with customers on various business engagements with OptiRisk.

 

Posted in News Analytics, Uncategorized, Workshops and Seminars | Leave a comment

Sentiment Analysis Applied to Finance

The growing importance of Sentiment Analysis applied to finance bring forth many questions to minds like why “Sentiment Analysis is important and why is online media and social media important to look at? Sentiment analysis tools automatically extract opinions, emotions and sentiments in text. It also allows individuals to get an opinion on reviews on a global scale. In terms of artificial intelligence, sentiment analysis uses a lot of mathematical calculation to determine whether the reviews capture positive or negative opinions.

Social media has created a new world of venting customer voice. Sentiment analysis offers organisations the ability to observe the different social media sites in real time and then act accordingly. Twitter is considered to be the ocean of sentiment data. Anyone can extract the data easily by searching on hash tags (#) on twitter.

I came across Michelle Price’s article “Taking sentiment analysis beyond trading” in Financial News on 3 December 2012. She said, “When the financial industry first embraced sentiment analysis of news and social media sites, it was chiefly regarded as a means of making money”.  She also referred to Richard Brown, head of quantitative and event-driven trading solutions at Thomson Reuters, saying: “In terms of the value from social media sites, we’re at the tip of the iceberg at what the industry can achieve.

 Sentiment analysis has developed as a technology that applies machine learning to make a rapid assessment of the sentiments expressed in news releases. Investment managers, funds managers, marketing managers, and equity investors are the direct beneficiaries of sentiment analysis technology.

News events impact market sentiment and financial news moves stock prices through a direct impact on a company’s expected future cash flows. Based on this concept UNICOM in collaboration with OptiRisk Systems is organizing a conference “Sentiment Analysis Applied to Finance”. This conference presents the current state of the art in the application of Sentiment Analysis to the respective models of trading, fund management and risk control. The conference presents in a summary form the research results in this fast-emerging field.

 Many leading technical specialists will participate in this conference and present overview talks, case studies of their fields and experiences.

UNICOM and OptiRisk have been working recently with a number of partner institutions who are highly regarded in the area of finance: PRMIA, the 14 – 10 Club @ the Royal Institution (The 14-10 Club is exclusively for finance professionals. It was set up in 2011 and is sponsored by Investment Banks and leading hedge funds). On the commercial side they also have partnership with The Mankoff Company < New York, USA >. In partnership with these professional institutions, UNICOM and OptiRisk are organising the “Sentiment Analysis Applied to Finance” conference in June/July to attract a wide audience.

Posted in News Analytics, Workshops and Seminars | Tagged , , , | Leave a comment

How to use R in Finance

OptiRisk Systems, in collaboration with UNICOM seminars, ran a very successful workshop “Finance with R” on 6 – 7 November 2012.

A significant number of people who were unable to attend the workshop requested us to repeat the workshop. We are delighted to advise that the “Finance with R” workshop will be running again on 5 – 6 March 2013. We hope anyone who have missed the November workshop will be able to attend. It’s a great opportunity to learn about using R in finance sector.

We have received positive feedback from the delegates on the recently concluded workshop. Attendees especially appreciated the topic “how to use R in Finance – Portfolio Optimisaiton and VaR”. Among the participants were Fitch Rating, HSBC, RMC, Solo Capital, UCL finance researchers, many PRMIA members and others.

The speakers discussed about the utilisation of statistical software R to analyse data in financial sector and presented real world examples. They showed the power of R, highlighted statistical models used in finance and gave fantastic presentations in this well articulated workshop.

Ronald Hochreiter, an R Practitioner will be presenting again in the forthcoming workshop in March. He was highly appreciated by all the delegates in the last workshop.

We look forward to your participation in the coming year, and hope that the power of R will soon be utilised fully across the Financial domain.

Posted in Workshops and Seminars | Tagged , , , , | Leave a comment

IBM Optimisation Forum

OptiRisk Systems in association with IBM UK and UNICOM Seminars, hosted a gala event on “Optimisation in the Financial Services Sector” On November 13, 2012 at the IBM South Bank Premises.

The event had participation from over 50 delegates representing Financial Firms, Asset Management Firms, Investment Banks as well as Academics.

The event participation had an interesting blend of industry and academia with reputed speakers like David Jessop (Head of Quantitative Research, UBS), Gautam Mitra (MD OptiRisk Systems, Professor Emeritus Brunel University) and Nalan Gulpinar (Professor, WBS) presenting.

The forum presented some cutting edge topics and challenged the thinking  of leading practitioners in using optimisation technologies in Asset Management and Portfolio Optimisation.

With the economy on slow road to recovery, and financial firms desperate to look for new ways to better manage their portfolios, this event was of paramount importance.

We do hope that in the coming months and years, some of these leading research outputs will be exploited commercially as banks will gradually identify the need to look for alternative ways to manage their risks

Here are the presentations from the forum:

Nalan Gulpinar

Gautam Mitra

Sofianne Oussedik

Sofianne Oussedik – Central Bank Case Study

David Jessop

Enjoy Reading!

Posted in Assest Liability Management, Portfolio Optimisation, Workshops and Seminars | Tagged , , , , , , , | Leave a comment

News Analytics: It’s about more than just fundamentals

In the first edition of Short n’ Sharpe, I would like to focus on the significance of News Analytics – a study of how news and information impacts the price formation, volatility and liquidity of a security or index (Mitra, OptiRisk, 2011) – in the context of financial markets and “tactical” (i.e. short-term, time-sensitive) portfolio allocations. In the process, we will also observe the crucial role of cognitive biases – apparently one of the principal deficiencies of the Efficient Market Hypothesis (EMH) – in the investment process, which in turn makes quantifying the extent of sentiment/momentum that much more crucial.

One might ask, is there a compelling reason to believe that systematic news analysis could help active market participants[1] effectively capture alpha[2]? Evidence suggests that the answer to that question may well lie in the affirmative; the answer is, in fact, embedded in frequent deviations of market valuations from underlying fundamentals until the news data-points are digested and reflected in asset prices. Put another way, it is through news analysis that we endeavor to disprove, albeit primarily in the near term, one of the EMH’s primary contentions – that “market participants [will] not be able to systematically profit from market inefficiencies” (The Efficient Market Hypothesis, Wikipedia).  Let’s consider a recent case where monitoring news-flow consistently and analyzing it methodically might have helped investors beat the market – that of how the politically controversial proposal of Foreign Direct Investment (FDI) in India’s multi-brand retail sector affected the share price performance and tradability of the country’s flagship department store owner, Pantaloon Retail.

I first go back to October 18th, 2011. This is when the Business Standard first reported the possibility of an FDI bill being introduced in Parliament as a standalone measure (Decision on multi-brand retail by year end, Business Standard ), thereby marking the inception of persistent speculation and news-flow around this issue. The proposal would be immensely positive for Pantaloon whose balance sheet held US$1bn in cumulative debt. As expected, trading volumes on Pantaloon’s stock spiked up as did its Implied Volatility, as shown in the charts below:

Pantaloon Retail – Share Price Action + Trading Volumes 2011 (Source: Bloomberg):

 

Pantaloon Retail – Implied Volatility (Source: Bloomberg):



Interestingly, however, despite the fundamentally positive nature of the proposal for Pantaloon’s balance sheet (and consequently its equity value), the stock collapsed by 19.82%. Why the divergence? This is where cognitive bias comes into play. Investors felt little confidence that the FDI bill, a standalone proposal at the time and a risk to India’s multiple small mom-and-pop shops, would be politically controversial and hence get opposed aggressively by socialist politicians in Parliament, then in session. Needless to say, the crowd was right. The bill was quashed and Pantaloon’s stock fell further post the revocation (FDI in retail – Opposition to corner govt in Parliament, Times of India). The underlying sentiment from a seemingly positive development, well, wasn’t as positive!

The bottom-line? An amalgamation of headline news and market/investor sentiment had impacted Pantaloon’s stock price, volatility and liquidity. As speculation surrounding this issue stayed elevated, so did implied volatility levels as indicated in the chart above.

Let us now fast-forward to September 14th, 2012. This is when the Cabinet finally approved the same measure along with many others. With Parliament not in session and the government’s principal opposing ally no longer a part of its coalition, there was little room for pushback and the bill was ratified into law immediately. Once again, a constructive outcome for Pantaloon, which now holds US$1.2bn of cumulative debt on its balance sheet (US$200mm more than a year ago). The positivity pertaining to the issue, unlike in October 2011, appeared more definitive and potent this time (FDI in multi-brand retail comes into effect, Hindustan Times). As in late 2011, Pantaloon’s trading volumes and volatility levels spiked up again. However, this time, since the announcement, the stock rallied, not collapsed, by 20% (37% to peak post the news) with the benchmark Sensex up 10% during the same period. Investors felt much more confident this time and, had they bought the stock, would have successfully generated alpha.

 

Pantaloon Retail – Share Price Action + Trading Volumes 2012 (Source: Bloomberg):

The above example explicitly showcases the role of cognitive biases in tactical investing, as well as how collective “sentiment” and its relevant magnitude (i.e. the relative extent of positivity or negativity, measured through linguistic readers and similar software; Thomson Reuters and RavenPack term such metrics “sentiment scores” and “news scores” respectively) have the knack of incorporating more than mere fundamentals in influencing and/or guiding asset prices (in our case, Pantaloon’s equity value).

The necessity of constantly quantifying the extent and direction of such sentiment or momentum, and its consequent impact on return/alpha-generation, is thus set to become an important aspect of portfolio management and allocation. In fact, this subject becomes that much more important when viewed against the backdrop of an exponentially higher share of algorithmic trading and Direct Market Access (DMA) pipes across financial markets globally (the subject of the next edition of Short n’ Sharpe).

For now, it seems safe to assume that [systematic] News Analytics – one among multiple modes of investing – might matter more than one thinks it does. How large a component it really becomes within the realm of portfolio management, though, will only be apparent over time.


[1] Long funds, hedge funds, principal investors and insurance companies.

[2] The excess return of an investment relative to its benchmark(s), adjusted for a given level of risk

Posted in News Analytics | Tagged , , , , , , , , , , , , , | Leave a comment

Finance with R

Finance with R
There are many possible statistical programs that can be used in financial research and applications. R has grown tremendously in recent years, both in terms of capabilities and users. There are more and more Finance applications used in R. R, in addition, runs on Macintoshes.
I myself attended a workshop “Finance with R” in November 2010 organised by OptiRisk Systems and got a wealth of new knowledge and tricks from more experienced professionals in this field.
If you have any interest in the statistical software R and want to know its use/applications in finance, then check this out Finance with R.
Anyone looking for knowledge of R software and how to get started its applications in finance will find this workshop invaluable.
The programme offers tips related to the following points:
• Preparations and installation of R
• Data import, data types and variables
• Simulations in R
• Graphics in R
• Exploratory analysis in R with special focus to time series data
• Applications in Finance: Portfolio optimisation and VaR
• R and Excel
• Preparing reports

Should you be interested then Brochure download

It provides the knowledge of the statistical software R for professionals and academics in finance. This is well articulated workshop and presents insight into the statistical models and concepts in R which are useful in various problems arising in Finance.
The presenters of this course are Ronald Hochreiter, an R Practitioner, and Patrick Burns, a specialist in statistical computing specializing in R. They use examples that are widely used in finance.

Posted in Uncategorized | Leave a comment