Getting Your Data Straight

I've made progress on the paper I'm working on. Unfortunately, this week has been a good illustration of a quote from McCloskey: I believe it went something like "90% of writing is getting your thoughts straight, and 90% of empirical work is getting your data straight."

Unfortunately, my data wasn't straight - I realized that I had used the wrong data code (a certain type of dividend distribution) from CRSP. So, my previous analysis was basically crap (that's a technical term for the unitiated) and had to be redone using the proper data set.

Luckily, it looks like my primary results after using the prper code, but with a few minor changes. For now, I'm still doing the preliminary descriptive stuff. Since I did the initial version of the paper in a hurry (hey - it was a conference deadline), I took a few shortcuts. This time, I'm going back to step 1 and going over every line of code, and (just as important), making sure I know how the sample changes at each point. As a result, I'm much more confident with my data this time around.

But doing the descriptive statistics is still (to me) about the most boring part of the paper. Still, it's gotta be done.

The Limits of Models

Here's an excellent piece on the Psi-Fi Blog, titled "Quibbles With Quants." Here's a choice part:

What the models failed to capture was that humans don’t behave in simple, predictable and uncorrelated ways. It’s impossible to overstate the importance of the way these models cope with correlation of peoples’ psychology. To sum it up: they don’t. Let me know if that’s too complex an analysis for the mathematical masters of the universe.

Anyone who’s ever been to a nightclub, a football game or even a very loud party will know that there are situations where we don’t act as individuals, buzzing about doing our own thing. These are occasions when we all suddenly stop being individuals and start doing the same thing – usually involving large quantities of drugs and some very bad singing. Although these sorts of events are specifically designed to trigger this behaviour – which is probably a deep evolutionary adaptation to sponsor group behaviour, useful when it comes to running down tasty antelope and dealing with giant, carnivorous sabre toothed beavers – it can also happen in other situations. Most stockmarket booms and busts are generated by similar group effects.

In general, people behave in an uncorrelated fashion right up until the point they don’t.

Read the whole thing here.

Momentum Effects and Firm Fundamentals

The more Long Chen's work I read, the more I like it. I recently mentioned one of his pieces on a new 3-factor model. Here's another, on the momentum effect, titled "Myopic Extrapolation, Price Momentum, and Price Reversal." In it, he links the well-known momentum effect to patterns in firm fundamentals. Here's the abstract:

The momentum profits are realized through price adjustments reflecting shocks to firm fundamentals after portfolio formation. In particular, there is a consistent cross - sectional trend, from short-term momentum to long-term reversal, that happens to earnings shocks, to revisions to expected future cash flows at all horizons, and to prices. The evidence suggests that investors myopically extrapolate current earnings shocks as if they were long lasting, which are then incorporated into prices and cash flow forecasts. Accordingly, the realized momentum profits can be completely explained by the cross - sectional variation of contemporaneous earnings shocks or revisions to future cash flows. Importantly, these cash flow variables dominate the lagged returns in explaining the realized momentum profits. As a result, the realized momentum profits represent cash flow news that has little to do with the ex ante expected returns. In fact, the ex ante expected momentum profits are significantly negative.

So, in essence, he finds that investors ignore mean-reverting patterns in firm earnings, and over-weight recent earnings shocks.

Very nice.

On an unrelated note, the Unknown Family will be traveling the next few days for a family reunion in West Virginia (the Unknown Wife's father grew up their, and that fork in the family tree has a get-together every year). So, unless I schedule a few pieces to post automatically, posting will likely be slim for the next few days.

Updating a Dataset Always Takes Longer Than Expected

I'm still working on updating my data. As usual, what I thought would be a "simple" three to four-day jobhas stretched out to almost two weeks of work. At least I'm not this person.

Happy 4th of July

We're just got back from a trip to Wal-Mart (what could be more American) to buy some clothes for the Unknown Daughter (she's grown enough that her old bathing suit no longer fits). The Unknown Wife and Unknown Daughter tried on clothes, while I wheeled the Unknown Baby Boy around the store until he went to sleep. Not surprisingly, the large-screen flat-panel TVs did the trick (based on initial indications, he's definitely a boy-child to the core).

Now we're getting ready to grill some critters, followed by fireworks. In the meanwhile, here are some links. They're from a previous year's post, but they.re worth repeating (after all at Financial Rounds, we're all about the efficiency thing):

The Declaration of Independence most people have never read it through. So take a few minutes and do so before going about your day.

Our Sacred Honor - a piece that recounts what happened to the signers of the Declaration

The Pledge of Allegiance - 'nuff said.

The Star Spangled Banner - the words to our national anthem and a video of Whitney Houston singing it.

Now go grill some meat, light some fireworks, and have a happy 4th of July.