Sun

Jun 8
2008

Robert Passarella

Robert Passarella

Equity Research in the Age of Web 2.0

As a new blogger to Radar and the program co-chair for Money:Tech 2009, I thought I'd introduce myself along with a topic we plan to present at the conference.

After spending many years on Wall Street watching the buy side (investing institutions) and working with the sell side (investment banks); all I can say is that I am excited with what the future holds. The Internet is quickly becoming 'the' vast store house of data, research and commentary that I hoped it would. If you think the Internet is for message board jockeys, blogs for late nights in pajamas, and social media for dating; you are missing point. If you want to add to your company and industry analysis, as well as, your investment process, you need to be an interactive user of the Internet.

So why are we, "the professionals", turning to the Internet? Or more specifically; why are we turning to web sites, blogs and social media for investment research, data and commentary? Well I believe there are a few reasons.

Just about every public company has some type of presence on the Internet. In certain domains, enthusiasts and experts are better informed than most analysts on the actual products, goods or services a company may provide. And they are sharing this knowledge and information with each other in open forums. And if you decide to graduate from lurker to participant the gains are even greater. An informed investment professional, that knows his tickers and sectors, can quickly digest this information into his investment process, just like his phone calls with a Gerson Lehman or Coleman Research expert. This type of high quality information can be found if you know where and how to look.

The emergence of low cost publishing tools, cheap web hosting, and various ways to monetize knowledge is another trend that can put some people on the same level as a CNBC anchor and others on equal footing with a top ranked analyst. Just think of my well known blogger friends like Money:Tech Chair Paul Kedrosky, Barry Ritholtz, Nouriel Roubini, and Roger Ehrenberg. Their collective readership and influence is greater than most sell side analysts.

The economics and dwindling company coverage of most sell side research shops has forced the buy side to scale and employ a do-it-yourself ethic. Your typical buy side analyst needs a place to turn for leads and information, especially when switching sectors and industries. Those places are quickly becoming SeekingAlpha, Wikipedia, LinkedIn and Facebook.

And the biggest reason of all; the established sources of main stream financial news and data are ubiquitous. Everyone on the buy side and sell side has Bloomberg, Reuters, and Dow Jones. And if everyone knows everything at the same time what advantage is that?

So where are we headed?

We are headed to a place where the tools, practices and techniques are starting to emerge. It's a place where anyone with a need and drive can learn to scrape data, do their own automated channel checks, while benchmarking and creating reference data. Have an idea; model it yourself, grab the data from Amazon, benchmark it against Yahoo Finance, and set up your filters in Google Reader for stories or articles related to your themes.

The key is to meld the old with the new. Mixing your custom new found data with established sources from your vendors to provide leading edge mashups that help you complete or flesh out a company's position within the competing forces - think Michael Porter on steroids.

Researching during the age of Web 2.0 doesn't mean blindly following rumors on a web site -- that way leads to madness & ruin. You need to think more like Ben Graham in the early days of modern Wall Street when speculation was the name of the game. Graham was able to find high quality sources of information that others thought to be a waste of time, and he put them to good use.

The same can be said for the tools at your disposal in the Web 2.0 era. Put them to good use and you will be rewarded.


 
Previous  |  Next

0 TrackBacks

TrackBack URL for this entry: http://blogs.oreilly.com/cgi-bin/mt/mt-t.cgi/6537

Comments: 13

  Alex Tolley [06.09.08 07:48 AM]

"The same can be said for the tools at your disposal in the Web 2.0 era. Put them to good use and you will be rewarded."

Maybe. The lure of getting information that is somehow not impounded into the market price is very seductive. Unfortunately it rarely works. Is there really any evidence that fund managers who have access to more or better information actually perform better? You could also drive yourself to distraction trying to assimilate the information flow.

The web has clearly made information much more readily available than it was before. Web 2.0 may even improve on that, although I would like to see concrete examples. But my sense is that any information that is web accessible will quickly become part of the price, indeed it has to eventually if it has any signal value.

  Martin Edic [06.09.08 08:02 AM]

The crowdsourcing of sentiment about investments will, if nothing else, give a clear indicator of how much attention a company or sector is getting which would be incredibly valuable. We're in the social media monitoring and analysis space so I have an obvious interest in this (although it has not been a market focus for us- may have to rethink that!) both for our business but also as an investor. I've been monitoring a few companies to see if it presages any moves. Too early to tell but pretty interesting...

  Alex Tolley [06.09.08 12:55 PM]

Martin: "The crowdsourcing of sentiment about investments will, if nothing else, give a clear indicator of how much attention a company or sector is getting which would be incredibly valuable."

For that you would have to show that you could reliably measure the crowd sentiment AND that this measurement was a reasonably reliable leading indicator of price movement. The market already has good coincident crowd sentiment indicators, e.g. buy/sell volume.

  Rob Passarella [06.09.08 02:00 PM]

Alex: I understand your sentiments about the lure of getting information before it is resident in the equity price and your feeling that it rarely works.

There is academic research out there and Dr. David Leinweber of Berkeley's Haas School did a great presentation at Money:Tech entitled "If you had everything computationally where would you put it financially" He addressed some of the issues upfront and provided cases and papers.

The examples do exist today even without the academic rigor. Case in point is Apple computer. Just think for a moment what most of us that cover Apple did today. We got our information on today's Keynote from blogs like engadget, gizmodo, and AppleInsider. Mobile Me as a story was broken weeks before the announcement today.
These sources are the ones following the story and accumulating the knowledge base. I can't tell you the quant ( you said the magic word signal) side, that is not my bag, but from a fundamental research POV, disregarding these sources is hazardous to your investment research process.

  Alex Tolley [06.09.08 05:23 PM]

Rob, I read Leinweber's case study on Accentia. While interesting, it is not convincing. If the information was available up to 14 months earlier, why had this not been impounded in the price soon after? But supposing you alone had been privy to that information and acted on it, your investment would have been "dead money" until the price surge triggered by the news story occurred.

I would also apply the argument against technical analysis too. Once web 2.0 is "discovered" by players, any advantage will quickly be lost.

Don't take my skepticism to mean that using the web is not a good idea. When I started as a sell side analyst in the dark ages of the mid 1980's in London, the information readily available on companies was so sparse that you could actually devote some time to building a scrapbook of newspaper clippings. getting useful information was usually done over lunches with the company directors, and even they had to be circumspect in what they could say. Today the rules (in the US) are that all material information must be made public very quickly. The ideal information to hold was private only for 1 day, allowing shares to be bought/sold before the market was aware of it. Today the wealth of information available is almost too much and could easily cause 'analysis paralysis'. Nevertheless, having good information at your finger tips is important to making an investment decision. I would simple argue the case has to be made to show that it can translate to superior returns.

  Abiel Reinhart [06.10.08 08:48 PM]

Rob,

I'm not very familiar with using online sources to do equity research, but I am familiar with sites that focus on the macro economy or on general business and finance themes (this includes a number of the sites you mention). One thing that is still lacking from many of these sites--but not all--is big numerical conclusions; for instance a hard numerical prediction about GDP growth, net losses from a certain class of mortgages, the total number of foreclosures over the next X years, the peak in oil prices, and so on and so forth. Without this type of analysis, I think it will be difficult for many online sources to have as much impact as more traditional analysts, even if they have a larger audience.

It seems that one of the major reasons why the blogs aren't publishing these types of detailed studies is that the blog model doesn't lend itself well to this type of study. With the blog model focusing so much on short and frequent posts, there is often just no time to do larger thematic pieces, particularly when they require the collection of raw data which will need to be cleaned and then joined together for use in statistical programs or visuals.

All that being said, somewhat longer pieces are being produced either regularly or on an occasional basis by such sites as Calculated Risk, RGE Monitor, and Econbrowser, with what seems to be positive results. Undoubtedly something that would help this become more widespread is if the cost of data collection was significantly reduced. As the cost of data collection falls, we can assume that a larger and larger number of online writers can perform their own calculations in a reasonable period of time.

  Shane Smith [06.11.08 12:16 AM]

Rob, it can be argued that web 2.0 has already moved beyond a role as source of primary research data, to dissemination of fully evolved (and freely available) research (check researchoracle.com) In theory, trading profits suffer from ubiquity/wide availability - but alternatively, when did investors begin to react homogeneously to the last thing read, in isolation? Your comments on this development would be interesting.

  Falafulu Fisi [06.11.08 08:46 AM]

Alex Tolley said...
Is there really any evidence that fund managers who have access to more or better information actually perform better?

Yes. Fund managers who use robotic algorithmic trading often beat the market on a regular basis. Read some of the referenced articles at the very bottom of the page on that link.

  Falafulu Fisi [06.11.08 09:03 AM]

Alex Tolley said...
But supposing you alone had been privy to that information and acted on it, your investment would have been "dead money" until the price surge triggered by the news story occurred.

Here is an excellent paper that described a text-mining system developed at CMU for financial news analytics, which they named Warren (Buffet) which mines important news that could potentially affect stock price movements and present them to users in realtime. Here is the abstract:

Abstract :
In this paper, we presentWarren, a multi-agent system for intelligent portfoliomanagement,
which is motivated by the great benefits of working in teams within the domain
of Distributed Artificial Intelligence (DAI) and TextMiner which takes advantage
of information retrieval techniques to complement quantitative financial information.
In the portfolio management domain, software agents that evaluate the risks associated
with the individual companies in a portfolio should be able to read news articles that
indicate the financial outlook of a company. There is a positive correlation between
news reports on a company’s financial outlook and its attractiveness as an investment.
Since it is impossible for financial analysts or investors to track and read each one, it
would be very helpful to have a technology for automatically analyzing news reports
that reflect positively or negatively on a company’s financial outlook. It is also necessary
for an agent to learn contextual changes in the news reports autonomously. To
accomplish these tasks, we devised a new text classification method and a sampling
method. With comprehensive quantitative information gathered by efficient coordinations
between agents, and the supplementing of quantitative information by financial
news analysis, we showed a successful application of amulti-agent system for portfolio
management.

The full paper (PDF) can be downloaded from the link below:
Financial News Analysis for Intelligent
Portfolio Management

Text mining is just starting in its application in the financial market news analysis in realtime, and there is no doubt that it will become the common battle field of fund managers in the near future to see who develops the best accurate financial analytic text-miner. The better capability of the textminer to read the market online sentiment it realtime, fund managers then could use those info to design a better strategy for trading.

  Falafulu Fisi [06.11.08 09:41 AM]

Robert Passarella, Quantitative finance is one of my domain of interest. I have just completed developing a computational finance API to be used in a financial market web analytics (realtime).

I see some web apps out there already, such as Covestor, Cake Financial and a few others, but I am not worry about arriving late to the party, since in the field of investment, there is only one thing that matters and that is how deep is the analytical capability of your system compared to your competitor. I see the other sites are still using traditional simple metrics, but the world of financial analytics had moved on from those simple methods, such as Sharpe ratio, Markowitz allocations, Maxdraw-down bench-marking, etc,... I am introducing the latest analytic algorithms which come straight from recent economic/finance literatures. Some of those techniques are unknown to even seasonal Wall Street analysts. In saying that, I do know that analysts from Goldman Sachs, Merrill Lynch , Credit Suisse and the likes, do adopt those latest techniques from the literatures. I know this since some authors of those papers that I had made contact with in the past requesting clarifications of their algorithms described in their papers, informed me of who had implemented those algorithms for commercial use.

My intension is to bring the same analytical techniques that only top notch analysts from Goldman Sachs & Credit Suisse, etc,..., to the ordinary & institutional investors.

One area, that I am keeping a tab on is the researches from the community of Econo-Physics (application of physics in Economics). These are physicists doing research in economics applying Physics principles. The field is still controversial to mainstream economics, but there are some cutting-edge discovery that originated from econo-Physics, such as the discovery that stock-prices don't follow the log-normal probability distribution, but in fact stock-prices do follow a power-law probability distribution.

There are some amazing researches from this community over recent years and econo-physicists such as that from Oxford Center for Computational Finance, which is one of the leading research center in this field.

I thought I had put away my Quantum Mechanics textbook in my bookshelf after my University days for good. I've just dusted it off recently to revise my memory over the subject again, since it would be vital in my algorithm development over the coming months.

By the ways Robert Passarella, have you had experienced from Wall Street? I am keen to have an informal chat (email) if you have, provided that you're interested.

  Pat [06.11.08 05:30 PM]

This looks interesting. Haven't read anything about it yet here in Germany.

  Fisher Investments [03.24.09 02:47 AM]

Interesting article about equity research. I work at Fisher Investments and MarketMinder is another great source for free information: Fisher Investments MarketMinder

  Ben J [05.14.09 03:52 PM]

I, too, work at Fisher Investments. We also have some additional information and a discussion forum on investment books found here: http://groups.fisherinvestments.com/investment-books

Post A Comment:

 (please be patient, comments may take awhile to post)






Type the characters you see in the picture above.

RECOMMENDED FOR YOU

RECENT COMMENTS