How to use Google search data to invest (podcast)

tradestream radio, discussing investing and technology

In my book, Tradestream, I talk a lot about what I call “Co-lateral Research”.  This is information inherently non-financial in its nature that investors can use to make better investment decisions.

Take Amazon Sales Ranking, for example.  Amazon provides almost real-time ranking of its best selling items.  While Amazon won’t reveal exactly how many units of Apple’s ($AAPL) iPad it’s selling, investors can get a qualitative feel for how well products are moving.

Summary

UNC Professor Joey Engelberg has been studying another form of co-lateral research, Google search data.  He’s been studying search trends for stocks (ie $PCLN or $NFLX) as a way to measure investor attention.  Prof Engelberg has found a linkage between changes in search volume and subsequent moves in stock prices.  He joins us for this installment of Tradestreaming Radio.

We discuss

  • which particular stocks investors pay attention do during the trading day
  • the demand side of news and information for stocks
  • how Google search volume is correlated to stock pricing
  • a trading strategy that uses search volume to beat the market

Listen below

Resources:

 

Differentiating your view on a stock

Last week, I interviewed Jim Valentine, author of Best Practices for Equity Research Analysts: Essentials for Buy-Side and Sell-Side Analysts. For professional researchers, Jim emphasized the need to have a differentiated view on a stock.

From our chat:

I think another thing that may be getting more into the investment arena, in terms of best practice, is to identify the two to four critical factors that impact every one of your stocks. This kind of goes back into the time management thing. What I find is that there are a lot of really smart analysts who know a lot about other companies, but they don’t have a differentiated view from consensus on any particular issues that are going to drive the stock. Ultimately they become a company analyst, rather than a stock analyst.

So the best practice, in effect, is to do some research, figure out what are those two to four critical factors, and then focus all your time on those for your companies, as opposed to all the other factors out there that are, in effect, noise.

Pro analysts vs. the rest of us

This need to differentiate makes sense for equity analysts: ultimately, the good ones get paid to pick spots when their information and view on an investment diverges from consensus.

But I’m torn if this makes total sense for the rest of us.   How important is it for the majority of investors to do something completely off the beaten path?  We know that stocks with greater investor recognition typically do better than those with less.  Do we really need to stick our investing necks out to try and find the next $GOOG or $AAPL?

In my own investing, I tried for years to come up with original stock picks — either ones that no one was looking at or ones that everyone hated on.  Now, I’ve automated much of my investment process leveraging the Tradestream and spend much of my time finding strategies that take me out of the investment selection process.

Where is the line between being an investing contrarian and not getting swept along with every fast-and-quick investment fad and just doing things according to “the book”?

More resources

Learn more about the book and Jim Valentine

Hedge funds beefing up after good returns, inflows in 2010

The hedge fund industry certainly took a drubbing in the bleak market years which were 2007 – 2008.  But, they’re back and they’re back on the heels of good performance and inflows in 2010.  Credit Suisse released its 2010 Hedge Fund Industry Review (.pdf)  today.

Few highlights:

  • Hedge funds, as measured by the Dow Jones Credit Suisse Hedge Fund Index, were up 10.95% for 2010 after posting positive performance for seven out of 12 months
  • On an asset-weighted basis, an estimated 81% of funds have surpassed previous high water marks as of December 31, 2010
  • The industry saw an estimated USD $8.5 billion in inflows for the fourth quarter, bringing overall inflows to $22.6 billion for the year. This represents the largest annual inflows into the space since 2007
  • The largest inflows in 2010 were seen in the Global Macro and Event Driven Sectors, up $16.8 billion and $13.9 billion respectively, while the largest outflows were seen in the Multi-Strategy sector which lost $16.9 billion
  • Including performance gains, current hedge fund industry assets under management (AUM) grew to $1.7 trillion as of December 31, 2010, up from $1.5 trillion on December 31, 2009
  • Research of returns from January 1996 through December 2010, indicates that smaller hedge funds (less than $100M AUM), have historically outperformed larger hedge funds (greater than $500M AUM) by 3.95% annually

Check out the whole report here.

PortfolioRunner: New portfolio analytics tool to leave beta

New portfolio analytics tool, PortfolioRunner, is readying to leave its beta testing and formally launch.  Well, it’s actually more than an analytics tool.  Portfolio Runner does 3 things:

  1. makes stock recommendations: using a variety of technical, fundamental and crowdsourcing algorithms, PR suggest additions stocks to buy/sell
  2. risk management: PR helps investors determine size of individual positions and suggests an exit strategy for each position
  3. news analysis: PR makes it easy for investors to follow breaking news on the fly, determining whether the news is positive or negative for the stock, and extracting relevant info.

Check it out and let us know what you think of it in the comments below.

When searching for stock gains, use Google (search data)

A wealth of information creates a poverty of attention

Smart investors avail themselves of all valuable resources as inputs into the investment research process.  I write about this faculty in my book Tradestream in the chapter “Co-lateral Research“.  What co-lateral research means is all the non-financial/non-traditional sources of information that can be used by investors to connect-the-dots.

I’ve written about Google Domestic Trends, search volume data Google has made public and overlayed on top of stock index charts.  GDT continues to be a good resource for investors.

And now, there’s more research to support using Google search data to auger where markets are headed.

In In Search of Attention, researchers found that Google’s Search Volume Index captures retail investors’ attention in stocks.

Among our sample of Russell 3000 stocks, stocks that experienced an increase in ASVI [me: abnormal search volume index reading] this week are associated with an outperformance of more than30 basis points (bps) on a characteristic-adjusted basis during the subsequent two weeks. This initial positive price pressure is almost completely reversed by the end of the year.

The paper also finds that increased search volume leading up to hot IPOs may be responsible for that big first-day pop! that such issues experience.

As the first paper that has really looked at search data from an investing standpoint, this should be piped and smoked.  In fact, the authors conclude the paper with a somewhat foretelling statement:

Search volume is an objective way to reveal and quantify the interests of investors and therefore should have many other potential applications in fi…nance. We leave those for future research.

Bring it on.

Source

In search for attention (Da, Engelberg, Gao), November, 2010

HT: Net//Worth

More commodity data/analysis available

Worldcrops.com just launched.  Headed by former Financial Times commodities editor Gary Mead, there’s more here than just commodity data porn.  Worldcrops.com is an offshoot of the VM Group, a UK-based advisory group focused on research and analysis of precious and base metals, energy, agribusiness, and renewables.

The site provides two differing levels of subscription — one (free) that gives access to raw data, news while the premium version provides a layer of analysis on top.

You can see the press release here.

Letter to the Google founder dudes

Wassup, Larry and Sergey.

Nice earnings last night.  26% growth, over $8b in the quarter.  Not too shabby.  I’m taking my $GOOG stock to the bank.

Oh, also, cool what you did to the “supposed” adult supervision afforded by (well, ex-) CEO Eric Schmidt.  It’s pretty sick how Larry just wrested control away from him and sidelined him to take back the company.  I’d give all my 24 hr access to fruitloops on the Googleplex to see the look on his face when you dropped that bomb on him.

For me, it didn’t really rock my world.  From the word on the Street (or at least on campus here), he never really had any control anyway. Like when you guys bought Android and didn’t even tell him.  That was awesome.

Eric may of had to go — I mean, with over 90% search dominance in so many markets, yet the inability to really monetize anything else and the makings of a Microsoft-type ($MSFT) European anti-trust inquisition mounting against the search firm, the company needed some fresh blood.

But this peep isn’t exactly sure that a return to its roots is what the firm needs.  With the growth in the company over the past few years (23k f’in Googlers, baby!), we’re kinda all over the place.  Add a lame attempt at buying Groupon — we need new leadership.  I mean, the last thing we want is to pull a Yang or Ballmer and bring back the lifeblood of our startup days — our founders — to steer these ships that have certainly outgrown their own abilities to manage such large organizations.

Oops — is that what just happened?  $YHOO’s Carol Bartz maybe-coulda-woulda been a better choice??  Maybe not.

Rock on,

Your faithful (and rich) Googler

Best way to trade the rumors? Bloomberg (and Tradestream) says short ’em

To a philosopher, all news as it is called, is gossip, and they who edit and read it are old women over their tea — Henry David Thoreau

Gossip is called gossip because it’s not always the truth — Justin Timberlake

With stocks, there is so much noise and pumping going on that investors can feel like they’re at a Motley Crue concert again.  So, how do investors using smart strategies and historical data profit from rumors?

Bloomberg is out with proprietary data today that suggests shorting stocks caught up in merger rumors is a viable, profitable strategy.

Electronic news services, brokerages and newspapers reported at least 1,875 rumors about potential buyouts of 717 companies between 2005 and 2010, according to data compiled by Bloomberg. A total of 104, or 14.5 percent, were acquired, the data show. While stocks that were the subject of takeover speculation initially jumped 2.9 percent, betting on declines yielded average profits of 1.2 percent in the next month, an annualized gain of 14 percent.

In Tradestream, I devote an entire chapter, Grind the Rumor Mill, to rumor mongering and how that plays out for investments – essentially short-selling a basket of M&A rumors.  This strategy works because while real acquisition targets see above-average appreciation, most rumored M&As don’t actually come to fruition.

I included a rumor model developed by Nudge’s Cass Sunstein that he used in his recent book, On Rumors: How Falsehoods Spread, Why We Believe Them, What Can Be Done (affiliate link).  This included identifying propagators, qualifying their prior beliefs, and predicting the cascading effect from any change/reinforcement of those priors.

Much of the guts and data behind this strategy was documented by Gao and Oler in “Rumors and Pre-Announcement Trading: Why Sell Target Stocks Before Acquisition Announcements?” (June 2008)

Data

The Strategy

  • Research: Scan the WSJ’s Heard on the Street for reported, but unsubstantiated merger and acquisition rumors
  • Adjust for market cap: The strategy works better when you remove companies with market cap >$20B
  • Short basket trade: Short sell a basket of these rumored targets and hold for 70 days after the rumor first appeared.  Cover.  Hedge if you like.
  • Timing best for hot M&A years: if M&A heats up (like now, right), the data show the strategy works even better

Last thing

The Bloomberg research found that this short-the-rumor strategy worked (+14%) even when it coincided with other contradictory bullish signals like call buying.

Call volume in New York-based Jefferies Group Inc. jumped amid unconfirmed takeover reports on Feb. 27, 2008. Calls on the company changed hands 12,692 times that day, 24 times the four- week average and the most in almost a year, and the shares gained 3.7 percent. A deal never occurred and Jefferies dropped 3.4 percent the next day, 10 percent the next week and 20 percent in 30 days. The S&P 500 lost 4.7 percent in a month.

Caveat emptor: I have not actually used this strategy in portfolios (I’m pretty much long only) and I think it would take balls of steel to really stick to it.

Further Reading on Investing and Rumors: