AmalgaMood’s Mood Index inverted on Friday May 31, 2013 and is now trending down.  This negative inversion negates the uptrend that began on February 9, 2012.

From February 9, 2012 to May 31, 2013, the S&P 500 index appreciated 21% and the S&P Global 100 Index appreciated 13%.  In other words, the markets have performed extremely well and have justified the positive call early last year.

At 2% from the all-time high (close), the S&P 500 index is currently seen by most as solidly within its bullish trend.  In fact, many market professionals have recently become bullish after being bearish for much of the past year.  A negative call at present appears out-of-consensus, but no more so than during previous negative inversion signals, which worked out favorably (though past performance may not be indicative of future results).

It should also be highlighted that after previous negative inversion signals, many global indices went on to make new marginal highs.  Positive price momentum can take markets higher even as the conditions that support longer term trends deteriorate.  In previous cases, these marginal highs were soon followed by significant equity market declines.  In other words, the Mood Index has historically inverted a bit early during negative inversions.

Using historical precedent, the downtrend should last from around 6 to 18 months.  During this time, general global conditions should deteriorate on the margin and work as a weight to global stock performance.

We’ll provide updates as the situation unfolds, stay posted.

Big Data is here is stay. Will people really stop tweeting, posting, or updating — in quasi real-time?

This should be a field day for everyone in the financial management sector. After all, we are always after more information, better information, more complete view of an investment case, real-time feedback, etc. Big Data offers this to us. But its appearance on the scene has occurred almost too quickly, and many in the financial market community are still struggling to understand what will be the impacts of Big Data on trading, equity analysis, investment strategy, compliance, and everything else related to financial management.

I personally have run into many colleagues (who maybe a year or two ago would have made fun of the concept of quantifying social media, news, and other information sources to create investment strategy) asking what does this all mean and how should they incorporate this mega trend into their proven investment analysis.

One tricky thing about this topic is that there are so many different / new skill sets that go into it. We are no longer talking about the things you might have learned in your MBA classes or even on a traditional trading desk. Discounted cash flow analysis is still useful, but it will not get you very far analyzing extremely large sets of data streaming in real-time. The new skill set includes, but not limited to, such disparate (and traditionally non-financial) things as:

• database architecture
• machine learning
• ontology creation
• sentiment analysis
• text analytics
• hadoop (apache ecosystem)
• parallel processing
• predictive analytics
• quantitative analysis
• social media monitoring

So, I have begun an interview series that explores these topics taping some of the leaders in these emerging fields.  Finance, like every other major sector, will be transformed by the impact of Big Data.  The goal of this interview series is to illuminate the path forward.

The Battle of the Quants conference in NYC, and particularly the panel which I moderated, was insightful on a number of levels.

The first significant insight is that we are in the midst of a period of fabulous disruption in the field of finance.  The arrival of Big Data to financial management is changing and will continue to change the landscape of the sector.  The days of traders, analysts, portfolio managers, economists, and strategists manually sifting through piles of information to find those few gems are coming to an end.  There is just too much data coming at too fast of a velocity from too many sources for such human-focused processes to sustain a competitive advantage.  Disruptive elements are bombarding the financial management field and there does not appear to be any slowing of this trend, if anything it appears to be accelerating.

The second significant insight is that fascinating innovations are emerging which will help to create new winners within the sector.  Expected innovations will sprout up in various areas within finance, but it appears that increasingly the consensus points to Big Data analytics as the core innovation area.  This term is rather broad, but in essence it transforms the disruptive element of too much data into manageable and quantifiable output.  For instance, streams of information coming in from Twitter or from the 24/7 global news cycle are analyzed using such things as sentiment tagging in order to make sense of trends that naturally overwhelm individual investment management professionals due to the amount and speed of the data.

In general, most tended to agree concerning the current flux of the sector, the disruption that Big Data has brought with it, and the innovations being produced leveraging Big Data.  From there, disagreements begin to appear.

The panel discussion which I moderated, “Analyzing, Tagging, Enabling, and Coding Sentiment Data for Use in Financial Models” produced a number of interesting disagreements of opinion.  The members of the panel agreed that quantitative financial models will increasingly include output from Big Data analytics, most specifically sentiment data (unsurprising considering that was the focus of the discussion), but disagreed on how the demand will materialize.

Bram Stalknecht from Semlab and Gary Fuchs from Digital Trowel upheld that future growth and innovation will come from allowing users (in this case quant hedge fund managers) the ability to create their own ontologies and from allowing users to modify how the text analytics portion of the analysis is conducted.  In short, they believed that end users will have the interest, time, and ability to master the text analytics portion of the process that today, for the most part, is outsourced to specialized firms.  End users will then be able to conduct more detailed and focused analysis – producing more original observations, which should result in greater differentiation from peers.

On the other end of the spectrum was Armando Gonzalez  from RavenPack.  He upheld that actually quant fund managers are more interested in receiving the numerical scores (in this case sentiment scores) from the third party specialist.  His point was that such investors will focus on their strengths, which include quant model building and not on text analytics, which requires a different skill set, experience, and training.

Future developments in this space, as well as new innovations concerning the financial management field in general, will be interesting to say the least.  I will provide my personal opinion on this issue in a future post.