Prediction Markets Election Contest
Over at the NYT's economics blog, Economix, David Leonhardt is running a prediction market contest, looking at odds of various Intrade contests. Pick any 3 of the 20 questions to answer, and the winner gets showered with untold glory and fame.
The contest has an interesting twist: Its based upon the odds at Intrade. That makes the choices less obvious and the strategies for winning the contest more intriguing. David asked me to submit a "guest" contest post for the Economix blog, which I did. You will find it below.
In addition to my post, David graciously invited all of you to participate, saying: "You have a lot of really smart readers. Let's see how well they can do in the contest." Challenge accepted. The contest runs til tomorrow morning at 6am. Go over to the prediction market contest and submit your best guesses.
Let me begin with this (weasely) caveat: While I have spent many years studying markets, I am not remotely a political analyst. In fact, I dislike politics, and I especially detest how it has polluted economics.
That said, be aware that my preferences in 2000 were the exact opposite of what actually occurred. My choices were McCain, Bradley, Gore, Bush – and that was pretty much the order they lost in.
And by way of full disclosure, I am probably best described as a Liberal Republican – low taxes, balanced budget, strong defense, no unnecessary overseas involvement, and no government involvement in personal matters (birth control, abortion, gay marriage, etc.) Liberal Republicans are now mythical creatures that no longer exist. I do not recognize the abomination that now calls itself the GOP. I guess that makes me an Independent.
OK, let’s move on to the contest, where whatever mathematical skill I may have could be of some small assistance in this contest.
My first thought: Picking the frontrunners and favorites appears to be a surefire path to the middle of the pack. To “win” this requires identifying which of the long shots really aren’t such long shots after all. In the markets, this is called variant perception – what the crowd (and the money they bet) betting thinks is an unlikely outcome, but is actually a higher probability result than most realize. (The best parallel company example is deep value stocks).
If this were actual money, I would most likely pick two favorites, and a single long shot. Something like:
9. North Carolina: Obama wins (1.9)
3. Obama’s electoral votes 379 or fewer (1.3)
11. Pennsylvania McCain wins (7.1)
[Tiebreaker] Winner popular vote share: 51.5%
That is the fiscally prudent thing to do, but that’s no fun. And since its not real cash, why not go for it?
What might be the variant outcome that most people think is very unlikely? I come up with 3: 1) A McCain victory; 2) A surprisingly tight race; 3) An Obama blowout. A McCain victory is (in my terrible political judgment) highly unlikely. The next variant perception scenario is either a very tight race, or an Obama blowout.
Looking at the questions, there are only a few that fit into this scenario. They are: Q2. Greater than 65%; 3. Obama’s electoral votes (out of 538) 380 or electoral votes more; 7. Obama wins Georgia; 10. Ohio McCain wins; 11. Pennsylvania McCain wins; 12. Virginia McCain wins; 13. Democrat-Republican breakdown 271 or more;
I don’t see McCain winning Ohio, Pennsylvania, or Virginia; He does have a good shot in Florida, but its not a great payoff (2.7 pts). Same with Obama winning Indiana (2.4 pts) and Georgia is a long shot.
That leaves the Obama blowout option the best odds (low possibility) relative to the points (highest points) So the picks I would make are:
2. Greater than 65% turnout;
3. Obama’s electoral votes (out of 538) 380 or electoral votes more
13. Democrat-Republican breakdown 271 or more.
[Tiebreaker] Winner popular vote share: 53.5%
While I give this a less than 20% chance of occurring, it’s the highest payout.
In an election where one of candidates ran many Hail Mary’s, it is only fitting.
One last caveat, so my clients don’t have a heart attack: This sort of low probability, high payoff is the exact opposite of how we run money in the office. It is why we’ve stayed out of trouble most of this year . . .
Prediction Markets and the Election: A Game
October 31, 2008, 6:42 pm
PBS Video: Taleb & Mandelbrot
Economist Nassim Nicholas Taleb and his mentor, mathematician Benoit Mandelbrot, speak with Paul Solman about chain reactions and predicting the financial crisis.
click for video
RAY SUAREZ: Finally tonight, we return to a subject on many minds these days: the financial crisis. Our economics correspondent, Paul Solman, checked back in with one particularly prominent voice in the investment world and his colleague, who guided his thinking.
Here is the pair's sobering conversation on what may lie ahead.
PAUL SOLMAN, NewsHour Economics Correspondent: One of the world's hottest investment advisers these days, Nassim Nicholas Taleb, author of "The Black Swan," who's been warning of a crash for years, betting on one, and winning big.
He's been ubiquitous in the financial media of late, from cable TV's "Colbert Report" to the BBC's "Newsnight," where he was infuriated by what he called "bogus accounting."
NASSIM NICHOLAS TALEB, Scholar and Author: The first thing I would get immediately, immediately, I would suspend something called value at risk, quantitative measures of risk used by banks, immediately.
PAUL SOLMAN: We sat down with Taleb and the man he calls his mentor, mathematician Benoit Mandelbrot, pioneer of fractal geometry and chaos theory. And even more than feeling vindicated, they're both scared.
NASSIM NICHOLAS TALEB: I don't know if we're entering the most difficult period since -- not since the Great Depression, since the American Revolution.
PAUL SOLMAN: The most serious situation we've been in since the American Revolution?
Top Theorists Examine Rippling Economic Turbulence
PBS, October 21, 2008
Alpha Into Beta
All About Alpha has a great Andrew Lo graphic depicting how Alpha eventually morphs into Beta:
A picture of the “betafication” of alpha
29 July 2008
What will happen to the quants in August 2017?
MIT and AlphaSimplex
Actual Merrill CDO Sale: 5.47% on the Dollar
An active trader pointed us to this very familiar looking off-balance sheet shenanigan found in the following paragraph regarding Merrill's CDO Sale.
Direct from yesterday's press release:
"On July 28, 2008, Merrill Lynch agreed to sell $30.6 billion gross notional amount of U.S. super senior ABS CDOs to an affiliate of Lone Star Funds for a purchase price of $6.7 billion. At the end of the second quarter of 2008, these CDOs were carried at $11.1 billion, and in connection with this sale Merrill Lynch will record a write-down of $4.4 billion pre-tax in the third quarter of 2008.
On a pro forma basis, this sale will reduce Merrill Lynch’s aggregate U.S. super senior ABS CDO long exposures from $19.9 billion at June 27, 2008, to $8.8 billion, the majority of which comprises older vintage collateral – 2005 and earlier. . .
Merrill Lynch will provide financing to the purchaser for approximately 75% of the purchase price. The recourse on this loan will be limited to the assets of the purchaser. The purchaser will not own any assets other than those sold pursuant to this transaction. The transaction is expected to close within 60 days."
Let's take this apart:
• Merrill appears to be moving $30.6 billion dollars of bad paper off of their books.
• This paper was carried at a value of $11.1, meaning there was almost $20B in prior related write downs.
• After this transaction, Merrill’s ABS CDO exposure in theory drops from $19.9 billion to $8.8 billion (hence, the $11.1B number).
• The $6.7B purchase price relative to the $30.6B notational value is 21.8% on the dollar
• Merrill is providing 75% of the financing –- and MER’s only recourse in the event of default is to retake the CDO paper back from the buyer.
• While Merrill hopes to be made whole, the reality is they still have potential exposure to these ABS CDOs via the financing;
• Actual sale price = 5.47% on the dollar
Less than five and half cents on the dollar? That's an even cheaper sale than originally advertised.
What this transaction actually accomplishes is getting the paper -- but not the full liability -- off of Merrill's books.
How very Enron-like !
We knew about the Merrill writedown on Friday… didn’t you?
FT, July 29th, 2008 at 11:46
Merrill's $5.7B Write-Down, $8.5B Share Issuance (July 2008)
Rinse. Lather. Repeat. (July 2008)
Merrill Lynch Announces Substantial Sale of U.S. ABS CDOs, Exposure Reduction of $11.1 Billion
Merrill Lynch Press Release, Monday July 28, 5:25 pm ET
My friend Paul has a chart from BP's data, Visualizing Global Oil Markets: 1965-2007 (not my favorite chart porn in the world).
Lately, I have been greatly enjoying the way numbers are depicted in a visual context, vis Visualizing Data. The site tends towards an eclectic depiction of data, which provides different ways of thinking about numbers, modeling, and the world.
I find it to be a fascinating exercise. This approach has very much colored how I approach and contextualize data. Some recent examples from the site:
Friday's WSJ front page article, Bernanke's Bubble Laboratory, is must reading:
"First came the tech-stock bubble. Then there were bubbles in housing and credit. Chinese stocks took off like a rocket. Now, as prices soar on every material from oil to corn, some suggest there's a bubble in commodities.
But how and why do bubbles form? Economists traditionally haven't offered much insight. From World War II till the mid-1990s, there weren't many U.S. investing manias for them to look at. The study of bubbles was left to economic historians sifting through musty records of 17th-century Dutch tulip-bulb prices and the like . . .
Now, the study of financial bubbles is hot . . . Among their conclusions:
Bubbles emerge at times when investors profoundly disagree about the significance of a big economic development, such as the birth of the Internet. Because it's so much harder to bet on prices going down than up, the bullish investors dominate.
Once they get going, financial bubbles are marked by huge increases in trading, making them easier to identify.
Manias can persist even though many smart people suspect a bubble, because no one of them has the firepower to successfully attack it. Only when skeptical investors act simultaneously -- a moment impossible to predict -- does the bubble pop."
Its now at the free section of WSJ.com.
Bernanke's Bubble Laboratory
Princeton Protégés of Fed Chief Study the Economics of Manias
WSJ, May 16, 2008; Page A1
Estimated Relative Standard Errors in Housing Data
Regarding yesterday's New Home Starts, an emailer writes:
You used to discuss the Commerce Dept.'s standard statistical error regularly. In light of that surprising Housing Start number, could you please update that?
Sure thing. I love this sort of data sifting exercise. (I used to do this all the time, but I could actually hear readers falling asleep through my screen).
Let's go to the Census Department release.
Privately-owned housing starts in April were at a seasonally adjusted annual rate of 1,032,000. This is 8.2 percent (±14.5%)* above the revised March estimate of 954,000, but is 30.6 percent (±6.7%) below the revised April 2007 rate of 1,487,000.
As is so often the case, the devil is in the details:
As far as the April Hosuing Starts go, a monthly change (seasonally adjusted annual rate) was 8.2%, versus an estimated relative standard error of ±14.5%. Hence, the monthly change is not statistically significant; that is, it is uncertain whether there was an increase or decrease in Housing Starts from March to April.
As to the 30.6% year over year drop -- that is ±6.7% -- and therefore is statistically significant.
[UPDATE: Flenerman in comments asks the same question]
And I thought I was the only one who cared about such mathematical trivialities . . .
NEW RESIDENTIAL CONSTRUCTION IN APRIL 2008
Manufacturing and Construction Division
U.S. Census Bureau, MAY 16, 2008 AT 8:30 A.M. EDT
Volatility Spike, parts III and IV
On Thursday, we noted the increase in volatility via a Financial Post column. Today's chart porn comes via the NYT & Barron's.
First up, the NYT, with this gorgeous info-graphic on volatility -- note the peak in 2002, which marked the bottom of the Bear markets (Oct 2002/March 2003):
Chart courtesy of NYT
Second, have a look at Dick Arms column in Barron's. Dick believes the recent volatility surge is a Bullish sign.
I have a lot of respect for Dick, as his methodology is statistically based and empirically driven.
Even if you disagree with him, you can at least respect his methodology, which has zero cheerleading content in it.
Chart courtesy of Barron's
Buy Volatility (January 14, 2006)
For Stocks, It’s the Wild West, East ...
NYT, March 29, 2008
Whiplashed? That's a Bullish Sign
Now Is the Time to Buy, Not Sell
RICHARD W. ARMS
Barron's, MARCH 31, 2008
Subprime Debt Values in a Recession
Amusing picture, via John Hussman:
Fun With Data Analysis: The Art of the Plausible
Barron's Mike Santoli looks at one of our favorite pet peeves: The use and abuse of data. His column this week, The Limits of History, notes that this has become especially prevalent of late:
"People who try to handicap the markets for a living practice the art of the plausible. Many trudge from conference room to lunch table to banquet hall lugging PowerPoint decks full of unobjectionable statistical touchstones for commission-wielding clients. At times of investor confusion and market dissonance, such as now, their art is often reduced to carving a slice out of economic history that ratifies their existing outlook."
Mike then proceeds to look at a "dog's breakfast of the kinds of historical analogies making the rounds." What is especially amusing about his list is the number of "Every time X happens, Y has occurred" that collectively produce all manners of mutually exclusive results. Since "X" occurred we will definitely have/avoid a recession; Stocks are undervalued/overvalued; Markets must rally/fall.
What is an investor to do? Whenever you are confronted with an "incontrovertible proof" based on historical data, prior to taking any action, I suggest asking yourself this short list of questions:
• Do we have enough historical examples? Is the data sample statistically significant?
• Causation or Correlation? Does "X" cause "Y" to occur? Or, are we presented with two things that may have the same underlying causes? Is there even interaction between X & Y?
• Coincidence? How possible is it that these two items are utterly unrelated (i.e., proof-we-are-clueless Superbowl indicator).
• Look for differentiating elements in different time periods: What factors are similar? What factors are different?
• Compare interest rates, inflation, dividend yield, P/E contraction or expansion, sentiment, overall market trend, business cycles -- across different eras. Might that account for potentially different outcomes?
• Any recent market environmental changes (regulation shift, financial innovation, etc.) have an impact? What might these specific changes do to the data? Consider Decimalization, ETFs, online trading, change in dividend tax, etc.
• Subjective versus Objective measures: Are the factors under discussion hard numerical data, squishy or somewhere in between? Percentage of stocks over 200 day moving average is objective; I find some chart pattern readings subjective. Earnings at time have been rather subjective; official inflation measures somewhere in the middle.
• Consider things in terms of probabilities, not outcomes: Assume a causative factor resulted in a specific event (X --> Y) 7 out of 9 times. The most you can say is that when "X" occurred in the past, it has resulted in "Y" approximately 78% of the time.
• There is a difference between historical occurrence and future likelihood. In the example above, this does not necessarily even mean that since "X" has just occurred, there is a 78% that "Y" will happen. Consider: was the first X/Y occurrence really a 100% or zero? Did the second one become 100% or 50%, then next a 66% or 33%?
• Contextualize data: Sometimes a single data point -- even a mean or median -- only tells half a story. Any data point can be trending or reversing. Going higher, lower, topping, bottoming. Each of these may have differing implications for what comes next. Inflation is high, but coming down. Gold is high -- and going up. It helps to think of data not as a still photograph, but as a frame in an ongoing film.
I'm sure there are more -- that short list is off the top of my head.
Any others? Please make your suggestions below. If we get enough good ones, I'll try to massage this into a more formal column.
The Limits of History
Barron's, MONDAY, JANUARY 28, 2008