John Tierney&#8217-s column the Science section of today&#8217-s New York Times discusses a method for forecasting difficult to predict events. The Copernican Method, advocated by Princeton physicist Richard Gott, allows one to generate confidence intervals that an event will occur using only the duration time until now (that is, how long the event has been at risk but has not occurred). Using the often not realistic assumption that there is nothing special about today, one can derive the ninety-five percent confidence interval for the time until the event occurs,

(1/39)*t_past &lt- t_future &lt- 39*t_past

where t_past is the duration time so far and t_future is the stochastic time until the event occurs.(*) Intuitively, events for which we have not observed a failure for a long-time are more likely to persist than ones which have only been in existence for a short time period. The article (along with the original Gott (1993) piece) give many examples of his formula at work such as how long Stonehenge will remain standing to how long political leaders will stay in power.

I remember reading the New York Times column in 1993 which first discussed this approach (sorry may be gated) and finding this to be not very convincing. Think about the Doomsday case. Of course today is quite different from the past: the events which could have led to man&#8217-s extinction in the past (largely exogenous natural events) are quite different from the dangers of today and the future (man-made events). But I always find data convincing. The NYT article claims that Gott made accurate forecasts of political tenure and the closing date of Broadway plays though I have been unable to track down the original predictions myself.

Well I doubt this will be of any use to folks investing in prediction markets. It has been about seven years since the last Democratic president. Applying Gott&#8217-s formula, this means with ninety-five percent accuracy we can say that the next Democratic administration will begin at least two months from now and no more than 273 years from now. I think we do not need a formula to figure that out.

(*) See Monton and Kierland (2006) for a derivation

Previous blog posts by Koleman Strumpf:

• Prediction Markets in the Classroom: Inkling Markets
• Slides of presentations from Conference on Corporate Applications of Prediction/Information Markets (1 November), Kansas City
• Summary of Conference on Corporate Applications of Prediction/Information Markets (1 November), Kansas City
• Reminder: Corporate Applications of Prediction Markets Conference (1 November)
• Conference: Corporate Applications of Prediction/Information Markets (Thursday, 1 November 2007)
• Win Justin’s Money? (re: Is there manipulation in the Hillary Clinton Intrade market? Redux.)
• Is there manipulation in the Hillary Clinton Intrade market?

That&#8217-s what tries to demonstrate event derivative trader &#8220-Vancheeswaran&#8221- in a comment on the Freakonomics blog post featuring a complacent interview of InTrade-TradeSports&#8217- John Delaney.

&#8212-

InTrade-TradeSports&#8217- John Delaney:

We listed a market on whether the U.S. Government would formally report that North Korea tested a missile in a certain manner. While the media reported that North Koreans did test a missile, it was not confirmed in an official U.S. Government release as was required in the market rules, so we settled the market according to the strict interpretation of the rules and not the understood intention of the market. This was understandably a real issue for some of our members and also for Intrade. It was a bad situation for everyone, really. We have learned from it.

Event derivative trader &#8220-Vancheeswaran&#8221-:

BULLSHIT.

Bryan Whitman ([email protected]) explicitly stated in print, in press conferences, and by email that North Korea fired multiple missiles into the Sea of Japan.

For instance, “North Korea fired a long-range Taepodong-2 missile and six short- and medium-range Scud and Nodong missiles. All landed in the Sea of Japan without incident.”

National Security Advisor Stephen Hadley, in a press conference at the White House, stated that the missiles “went out about 275 miles” into the Sea of Japan.
http://www.whitehouse.gov/news/releases/2006/07/20060704-1.html

There are many other examples of the U.S. military and government (not just the press) confirming that the missiles were launched and approximately where they landed.

John Delaney is completely misrepresenting what happened, just as he did at the time of the launches.

As I wrote last week, I will have my say on this Freakonomics interview. Stay tuned, folks.

In “Is Manipulation Good for a Prediction Market,” Eric Zitzewitz seeks to counter my view that “Deep Pocketed Manipulators are a Prediction Market’s Friend” by suggesting that even if my argument is correct, manipulators still may be a net bad for prediction markets.

I argued that a manipulator ends up subsidizing the participation of informed traders in the market by putting money on uneconomic positions. Eric responded with three reasons that manipulation nonetheless may tend to undermine prediction markets. In brief (and in my words), he said:

1. People have a distaste for behaviors which violate the spirit of the activity.
2. The possibility of manipulation, and therefore a biased indicator, will lead decision makers to rely on less efficient methods for information gathering.
3. Manipulation makes it harder to understand what is happening in the market – manipulation is harder to decode than noise – because the manipulator actively tries to trick the market.

I suspect Eric is right that many people will instinctively oppose manipulation– even when a manipulator&#8217-s actions comply with the explicit rules of the market, the conduct seems to violate the spirit of the enterprise. The manipulators aren&#8217-t betting that they can out predict you, they don&#8217-t care about out-predicting you, they are just using the market in the attempt to buy a signal for which they have a private value. This violation of the spirit of the enterprise probably strikes most people as just wrong.

I think the best response to instinctive opposition to manipulators is &#8220-get over it&#8221-. Hold your nose if you have to, but take as much of the manipulator’s money as you can. (If it troubles you to take advantage of manipulators this way, consider the public good you are doing by imposing a tax on undesirable behavior.)

Another obvious and useful response is to become more skilled at reading market prices, so as to not be readily fooled. In the face of a dramatic price change, consult other exchanges or the prices in related markets in order to gain context. Deploying such sophistication increases the cost and diminishes the usefulness of price manipulation.

The second and third points Eric makes suggests that the possibility of manipulation will reduce the value of prediction markets and lead decision makers to rely on noisier, but less manipulable tools for information gathering. (E.g., use expert forecasts instead of manipulable prediction markets.) In these cases, manipulation leads to real economic losses.

Useful responses here constitute a research program for the academics and market developers: what effects do various rules have on the potential for manipulation? what effects do the rules have on market efficiency? can manipulation be detected by analysis of trading patterns (perhaps something a prediction market could implement itself), or would more extensive investigations be necessary (likely beyond the capacity of a prediction market)?

Prediction markets, when they work well, create something of value: a price that can serve as a useful probability forecast of future events or conditions. If anyone relies on these prices to make decisions of economic significance, that reliance will create incentives for third parties to influence the prices. On these grounds I think the potential for manipulation is inherent in the enterprise.

So what? To some extent the natural balancing effects of the market will tend to counter manipulation, because manipulation provides a subsidy to participation by informed traders, and informed trading counters manipulation. This process is unlikely to be transparent, however, given that manipulation only works if it is disguised. The best response to the residual prospect of manipulation seems to be to resort to informed-rather-than-naive interpretation of prediction market prices, and research into market rules which may enhance usefulness of prediction markets.

The top 10 most active prediction markets at HedgeStreet:

Contracts – Time remaining – Underlaying Delayed Price – Bid – Ask – Last – Change
4PM $100 GBP/USD &gt- 2.0075 4h 13m 2.01370 95.00 100.00 95.00 29%
4PM $100 EUR/USD &gt- 1.3575 4h 13m 1.36240 95.00 100.00 89.00 34%
2:30PM $100 Crude Oil &gt- $70.00 2h 43m 69.67 28.50 35.50 29.50 -65%
2:30PM $100 Crude Oil &gt- $69.00 2h 43m 69.67 95.50 100.00 96.00 -1%
$100 Halliburton EPS &gt- $.56 18 days 0.56 76.50 96.50 81.50 -8%
4PM $100 USD/CHF &gt- 1.2150 4h 13m 1.21145 – 8.00 35.00 12%
2:30PM $100 Crude Oil &gt- $69.50 2h 43m 69.67 77.50 79.00 78.00 -16%
2:30PM $100 Crude Oil &gt- $71.00 2h 43m 69.67 – 5.00 6.00 -77%
1:30PM $100 Gold &gt- $657.50 1h 43m 659.6 87.00 91.00 81.50 234%
4PM $100 EUR/USD &gt- 1.3625 4h 13m 1.36240 52.00 66.00 18.50 64%

Technical Note: EPS = Earnings Per Share

&#8212-

HedgeStreet organizes many EPS prediction markets. Here&#8217-s the contract for the Google EPS event derivative excerpted from the HedgeStreet site:

Google EPS

Summary

Google®1 EPS Binary contracts allow traders to take a particular view on the &#8216-Earnings per Share&#8217- number (EPS) reported by Google, Inc. (&#8221-Google&#8221-) for a specific fiscal calendar quarter. At issue is the quarterly EPS number (defined as the net income per common share of stock, non-diluted basis) released by Google for the relevant fiscal quarter. A list of upcoming release dates can be found by visiting the Google website.

Example

Google EPS &gt- $3.55 (18 Jul 07)
This Binary allows traders to take a position on whether the EPS number reported by Google will be greater than $3.55 on July 19, 2007.

Asset
&#8220-Google EPS&#8221- specifies that the underlying for this Binary is the quarterly EPS reported by Google (in US dollars).

Strike Price
&#8220-$3.55&#8243- specifies that payout for this Binary is based solely on quarterly EPS for Google reported on the settlement date is greater than $3.55.

Last Trading Day
&#8220-(18 Jul 07)&#8221- specifies that this Binary will last trade on July 18, 2007.

Position
Buy if you think the quarterly EPS for Google reported in the company&#8217-s earnings release on July 19, 2007 will be greater than $3.55.

Sell if you think the quarterly EPS for Google reported in the company&#8217-s earnings release on July 19, 2007 will be less than or equal to $3.55.

Payout
Buyers make $100 if the quarterly EPS for Google reported in the company&#8217-s earnings release is greater than $3.55 on July 19, 2007.

Sellers make $100 if the quarterly EPS for Google reported in the company&#8217-s earnings release is less than or equal to $3.55 on July 19, 2007.

Definitions

Binary Name Format
[ASSET][STRIKE PRICE][LAST TRADING DAY][POSITION]

Asset
&#8220-Google EPS&#8221- specifies that the underlying for this Binary is the quarterly EPS reported by Google (in US dollars).

Strike Price
Specifies that the payout for this Binary is based solely on whether the quarterly EPS for Google reported in the company&#8217-s earning&#8217-s release on the settlement date is:

&#8220-&gt- $X&#8221- -greater than a specific level, $X.
&#8220-&lt- $X&#8221- -less than a specific level, $X.
&#8220-$X to $Y&#8221- -greater than or equal to $X and less than or equal to $Y (assuming $X &lt- $Y).
&#8220-= $X&#8221- -equal to a specific level, $X.

Last Trading Day
&#8220-(DD MON YY)&#8221- specifies that this Binary will last trade on a specific date, MON DD, YYYY.

Position
Buy if you think the quarterly EPS for Google reported in the company&#8217-s earnings release on the specified settlement date will be greater than the strike price.

Sell if you think the quarterly EPS for Google reported in the company&#8217-s earnings release on the specified settlement date will be less than or equal to the strike price.

Payout
Buyers make $100 if the quarterly EPS for Google reported in the company&#8217-s earnings release is greater than the strike price on the settlement date.

Sellers make $100 if the quarterly EPS for Google reported in the company&#8217-s earnings release is less than or equal to the strike price on the settlement date.

Trading Conventions

Trading Hours
The regular trading hours for Google EPS Binaries is from 8:00 am (ET) to 4:00 pm (ET). The trading of the Google EPS contract is halted at 4:00 pm (ET) on the last trading date.

Regular Trading Days
Google EPS contracts cannot be traded on their settlement date. Otherwise, listed Google EPS contracts can usually be traded any day HedgeStreet is open (see the HedgeStreet Calendar).

Last Trading Day
Google EPS Binaries will stop trading as of the date listed in the Binary&#8217-s name and the &#8220-At a Glance&#8221- table (right). The usual last trading day for Google EPS Binaries is one business day prior to the date on which Google has scheduled to release its Quarterly EPS for the particular fiscal quarter.

Settlement Date
Google EPS Binaries will settle as of the date listed in the &#8220-At a Glance&#8221- table (right). The usual settlement date for Google EPS Binaries is the day of the scheduled Google earning&#8217-s release. HedgeStreet reserves the right to postpone this settlement date if the release of the expiration value is delayed.

Settlement Time
Google EPS Binaries will settle within an hour of 4:00 pm (ET) on the settlement date.

Expiration Value
The expiration value is the number used to determine the settlement value for this Binary. It is the level of the Quarterly EPS of Google, Inc. (&#8221-Google&#8221-) for a specific fiscal calendar quarter, measured in U.S. dollars, as reported by Google in an 8-K Report filed with the Securities and Exchange Commission (&#8221-SEC&#8221-) or, if no such report is issued, as reported in its Quarterly 10-Q Report or Annual 10-K Report, as applicable, filed with the SEC.

For a detailed summary of the Source Agencies providing the expiration values for all Binaries, please refer to the Settlement Source table.

Settlement Value
The settlement value for Google EPS Binary contracts is either $0 or $100.

Speculative Position Limits
There are currently no Position Limits for Google EPS Binary contracts.

Halted Markets
In the event that any market irregularities are declared by HedgeStreet, trading in this Binary market may be halted. If it is determined by HedgeStreet that the market must be halted for any other reason, an explanation will be posted on HedgeStreet&#8217-s website within a reasonable amount of time but no later than 24 hours after the initiation of the halt.

Addition of New Binaries
To provide as extensive and relevant a market as possible, HedgeStreet may add new Binary offerings frequently and at its sole discretion. In some cases, new Binary offerings may affect the demand (and thus in some cases the liquidity and/or trading price) for existing, related Binaries. Traders should consider this dynamic aspect of HedgeStreet when making trading decisions.

Strategies

You can use Google EPS Binaries to speculate.

Speculate
Speculating involves taking a view on the outcome of an economic event with the hope of making a profit.

-Do you feel you have an insight into the future performance of Google?

-Would you like to speculate based on your belief about whether Google EPS will rise, fall, or remain the same?

Conventional investment markets don&#8217-t provide an opportunity for retail investors to speculate on a firm&#8217-s earnings announcements. Binaries let you invest the amount you choose on either increases or decreases in Google EPS.

FAQ

What is Google?
Google, Inc. (&#8221-Google&#8221-) is an American public corporation, specializing in Internet searching and online advertising. The company&#8217-s products and services are Google.com that offers Google Base, Google Video and YouTube. In addition, the firm&#8217-s offering also include: Gmail, Blogger, Google Docs and Spreadsheets, Google GEO (offering earth and local maps), as well as a plethora of other products. Google was founded in 1998 and is headquartered in Mountain View, California.

What are Earnings per Share?
EPS is the portion of a company&#8217-s profit allocated to each outstanding share of common stock, serving as an indicator of a firm&#8217-s profitability. The Financial Accounting Standards Board (FASB) requires companies&#8217- income statements to report EPS for each of the four major categories of the income statement: continuing operations, discontinued operations, extraordinary items, and net income. EPS for Net Income is calculated as:

(Net Income – Dividends on Preferred Stock)/Weighted Average of Common Shares

Why trade Google EPS?
Earnings results of a public company, including both Earnings per Share (EPS) and Revenue, provide insight into the financial health of such company and provide information on the strength of the industrial segment at large. Earnings results are not themselves securities, although they can influence security prices. There are other key measures of corporate progress, though, as well as other outside macro economic events that also influence the value of a given security. Google EPS Binaries allow traders to speculate on whether Google EPS will increase, decrease, or remain the same (as compared to previous quarter&#8217-s release or various analyst estimates).

What is the Financial Accounting Standards Board (FASB)?
Since 1973, the Financial Accounting Standards Board (FASB) has been the designated organization in the private sector for establishing standards of financial accounting and reporting (US GAAP). They are officially recognized as authoritative by the Securities and Exchange Commission (SEC) as the organization responsible for setting accounting standards for public companies in the U.S. The FASB&#8217-s mission is &#8220-to establish and improve standards of financial accounting and reporting for the guidance and education of the public, including issuers, auditors, and users of financial information.&#8221-

What is the Generally Accepted Accounting Principles (GAAP)?
Generally Accepted Accounting Principles refers to a set of accounting principles, standards and procedures that companies use to compile their financial statements set by the FASB. Every country has its own standard accounting practice version of GAAP with standards set by a national governing body. US GAAP is not written in law, although the U.S. Securities and Exchange Commission (SEC) requires that it be followed in financial reporting by publicly-traded companies.

What is the Securities and Exchange Commission (SEC)?
SEC is a U.S. government commission created by Congress to regulate the securities markets, protect investors and monitor corporate takeovers in the U.S. The SEC requires public companies to disclose meaningful financial and other information to the public. This provides a common pool of knowledge for all investors to use to judge for themselves whether to buy, sell, or hold a particular security. The SEC is composed of five commissioners appointed by the U.S. President and approved by the Senate. The SEC is designed to promote full public disclosure and to protect the investing public against fraudulent and manipulative practices in the securities markets.

Where can I find the most recent Google EPS release?
The Google EPS information is released each quarter and available on the Google or SEC website.

What is the Expiration Value?
The Expiration Value is the Quarterly EPS (net, non-diluted) of Google, Inc. (&#8221-Google&#8221-) for a specific fiscal calendar quarter, measured in U.S. dollars, as reported by Google in an 8-K Report filed with the Securities and Exchange Commission (&#8221-SEC&#8221-) or, if no such report is issued, as reported in its Quarterly 10-Q Report or Annual 10-K Report as applicable, filed with the SEC.

Where can I find background information related to this instrument?
For general EPS background information, we recommend the following:

* Investopedia.com: EPS
* Bloomberg.com Earnings

**The data displayed herein represents certain types of contracts offered on HedgeStreet that have an underlying expiration value based on the quarterly Earnings per Share or Revenue of publicly traded companies listed on a national securities exchange. The data is being provided for informational purposes only and should not be the sole information relied upon in making investment decisions. HedgeStreet does not guarantee or make any representation as to the accuracy and completeness of the information provided.

1 Google® is a registered trade mark of Google, Inc. HedgeStreet, Inc. is not affiliated with Google, Inc. and neither Google, Inc., nor its affiliates, sponsor or endorse HedgeStreet, Inc. in any way.

Next: HEDGESTREET: Earnings Per Share Prediction Markets + Merger And Acquisition Prediction Markets

Previous blog posts by Chris F. Masse:

• The CFTC is going to close the comments in 11 days. We have 11 days left to convince the CFTC to accept FOR-PROFIT prediction exchanges, and counter the evil petition organized by the American Enterprise Institute (which has on its payroll Paul Wolfowitz, the bright masterminder of the Iraq war).
• The definitive proof that FOR-PROFIT prediction exchanges (like BetFair and InTrade) are the best organizers of socially valuable prediction markets (like those on global warming and climate change).
• Fairness Doctrine prediction markets
• 2 MILLION TRADES LATER: Inkling’s play-money prediction markets are accurate —too.
• Web Forums on Prediction Markets
• Jason Ruspini will answer SOME of these CFTC questions. — 12 days left, Jason.
• QUIZZ OF THE DAY: Which blog is the most open minded?

Note the plural at &#8220-predictions&#8221-.

FAQ:

What does &#8220-short&#8221- and &#8220-cover&#8221- mean?
You know the maxim &#8220-Buy low, sell high&#8221-? Well, if you think the price of a proposition will go down, not up, you can &#8220-short&#8221- that prop and make money as it falls. Here&#8217-s how it works: When you short a prop, you borrow shares and sell them to another buyer. Eventually you must repay, or &#8220-cover,&#8221- the shares you&#8217-ve borrowed, by buying more shares at the new (and hopefully lower) price and returning those shares to the lender. Your net profit is the difference in price between the shares when you &#8220-short&#8221- them and when you &#8220-cover&#8221- them. So not only can you buy low and sell high, you can short high and cover low.

The answer to the proposition I&#8217-m interested in won&#8217-t be decided until 2050. What gives?
Long-term propositions are an important part of the market. Even though the exchange in its current form (or the Internet in its, for that matter) may not even be around in 50 years to see a long-term prop finally pay out, you can still profit greatly from trading it because its price will always serve as an indication of whether or not the market thinks the proposition will eventually come true. Take the &#8220-Will Androids Defeat a Team of Humans in Soccer by 2050?&#8221- [link] proposition as an example. We won&#8217-t know the answer to this one for quite some time, but if next week, Honda were to demo its humanoid robot ASIMO&#8217-s amazing new ability to run and kick a soccer ball at the same time, the price of this stock is probably going to jump. And if you&#8217-re holding shares, you&#8217-ve just made some money.

What&#8217-s a limit order?
A limit order is a way to automatically buy or sell a given number of shares of a proposition based on its price, allowing you to make smart trades at the right strategic moment without having to keep an eye glued to the market at all times.

Let&#8217-s say a proposition you&#8217-re interested in has been fluctuating between POP$50 and POP$75. If you know you want to buy, but only when the price reaches the low end of where it&#8217-s been fluctuating, you can set a limit order for 100 shares at POP$55. This means that the moment the stock&#8217-s price drops a penny below POP$55, the market will automatically buy 100 shares of the stock for you. Same goes for selling: If you wanted to sell your shares near the high end, you could set a limit order to sell 100 shares when the price reached POP$70. A limit order&#8217-s expiration date governs how long it will remain in effect-anywhere from one day to one month.

Previous blog posts by Chris F. Masse:

• 24 hours after the launch of the “Prediction Markets” group at LinkedIn, we have already 39 members —both prediction market luminaries and simple people (trading the event derivatives or collecting the market-generated probabilities).
• That was ubber world star Barack Obama in Berlin, during his July 2008 speech at the Victory Column. Spot all the digital cameras pointing to the socialist Messiah. Snatching something to bring at home — “see, I was there”.
• If you want your affiliation with the “Prediction Markets” group to appear on your LinkedIn profile, then click on “Edit Public Profile Settings”, and check the “Groups” option.
• If you want to connect with InTrade CEO John Delaney on LinkedIn…
• Do join the “Prediction Markets” group at LinkedIn, if you have a strong interest in the prediction markets or if you work in the prediction market industry. It’s free, and that’s a way for the LinkedIn visitors browsing stuff about prediction markets to stumble upon your resume / profile.
• You can now join the LinkedIn group on Prediction Markets.
• Nigel Eccles says that HubDub generates “data on peoples’ reputations for accurately analyzing and forecasting future events”.

Bruke Hansen:

[…] The day concluded with three papers covering prediction markets, the first by Robin Hanson, the world&#8217-s leading proponent of prediction markets and former head of DARPA&#8217-s notorious &#8220-terror casino&#8221-. Robin Hanson – futarchist, cryogenic enthusiast, and all around Kool Aid-chugger – is the son of Baptist preacher, and he brings to the cause an eerie religiosity. A former physics student, Hanson admits that he got a PhD in economics so people would take his ideas seriously: truly a cart-before-the-horse moment. One of the original attacks on the &#8220-terror casino&#8221- was that the market was too thin not to be manipulated, and could steer policy in the wrong direction. So what would a true believer do? Of course, Hanson&#8217-s paper claimed that market manipulation actually makes markets more accurate, not less, by increasing the so-called &#8220-noise&#8221- of the market. Of course, like virtually all the presentations at the conference, this was concocted as nothing more than a laboratory experiment. Real statistics from futures exchange Intrade, for example – a conference sponsor, no less – were few and far between.

Totally unfair to Robin Hanson.

#1. Robin Hanson is indeed the world&#8217-s &#8220-reigning expert of the field of prediction markets&#8221-&#8230- FOR ALL THE GOOD REASONS NOT CITED BY THE REGISTER. See the Prediction Markets Timeline for more information.

#2. Robin Hanson is overly optimistic on his concept of &#8220-decision markets&#8221-, that is, complex conditional prediction markets used as a decision tool. Robin Hanson is wrong on the fact that there is a need for market-generated decisions, because the executives and politicians will never be willing to let a machine decide for them. That said, it&#8217-s not much of a problem because all Robin Hanson&#8217-s work on &#8220-decision markets&#8221- can be recycled for the &#8220-decision-aid markets&#8221-, that is, the step above the classic internal prediction markets that we know today.

#3. Yes, Robin Hanson can be out of whack sometimes, but, at the contrary, that&#8217-s a plus: Craziness and creativity go hand in hand. Nobody gives the first fig about what Robin Hanson does on his free time. What we see, in the field of prediction markets, is that he thinks outside the box, being an new institution designer, for the benefit of us all.

#4. The comment on the 2000–2003 US DoD&#8217-s DARPA&#8217-s IAO&#8217-s FutureMAP/Policy Analysis Market project is totally misinformed. The PAM project was sound. The only (very minor) valid criticism to make is that Robin Hanson insisted, wrongfully in my view, to focus on the Mid-East, where it&#8217-s well know that the American traders have few advanced indicators. Other than that, let&#8217-s not forget that PAM gave birth to a new technology, MSR, used today in three public prediction exchanges (Inkling Markets, WSX, and Media Predict) on top of being used at MicroSoft Research and by all Inkling&#8217-s clients.

#5. As a blogger, let me say that Robin Hanson is good material for me. I can sometimes laud his ideas, and at other times ridicule his ludicrous views. Sometimes, I&#8217-m wrong, and at other times, I may score a point against him. I don&#8217-t have this liberty with Justin Wolfers, for instance. The only thing I could criticize in Justin Wolfers was his Australian accent, but I got a boo from my audience for picking him as a target.

Previous: The Growth of Gambling and Prediction Markets: Economic and Financial Implications

NEXT: The truth on the 2000–2003 US DoD’s DARPA’s IAO’s FutureMAP/Policy Analysis Market project.

This is an edited version of a post on pancrit.org commenting on the public letter advocating safe harbor for small-scale academic prediction markets. I can see why they limited their goals as they did, and I agree that everything they advocated should be legal, but I think they may have limited their objectives just enough to prevent any big wins.

One thing that Chris Masse seems to constantly argue is that Prediction Markets on dry subjects need to be accompanied by entertaining questions in order to to keep the audience&#8217-s attention. The economists had good reasons for shying away from recommending that sports betting should be included, but there are many other topics that diverse markets could include that give traders a reason to check back in. The range from the obvious entertainment questions (movie earnings and oscar winners) to legislative outcomes (bills passing and control of particular legislative bodies) and introduction and market success of new technologies. While these kinds of questions might be out of place on some single-topic markets modeled after the University of Iowa&#8217-s markets on elections, the internal corporate markets that they also mentioned often use them to help maintain interest. The letter&#8217-s recommendations that the CFTC &#8220-allow contracts that price an economically meaningful risk or uncertainty&#8221- unnecessarily limits the kinds of contracts that would be allowed.

Back on the side of supporting the letter&#8217-s authors again, I&#8217-d have to admit that if the CFTC or Congress acts to implement anything resembling the recommendation it would very likely increase Prediction Market activity greatly, and eventually lead to a broader acceptance. If the initial definition is too narrow, however, questions that don&#8217-t have clear economic implications (in the view of Congress and the regulators) might be stuck offshore for a long time to come.

Statement on Prediction Markets – (Click here to read the abstract and download the petition from the SSRN site) – by Kenneth J. Arrow, Robert Forsythe, Michael Gorham, Robert Hahn, Robin Hanson, Daniel Kahneman, John O. Ledyard, Saul Levmore, Robert Litan, Paul Milgrom, Forrest D. Nelson, George R. Neumann, Charles R. Plott, Thomas C. Schelling, Robert J. Shiller, Vernon L. Smith, Erik Snowberg, Cass R. Sunstein, Paul C. Tetlock, Philip E. Tetlock, Hal R. Varian, Marco Ottaviani, Justin Wolfers, and Eric Zitzewitz – 2007-05-XX

&#8212-

Executive Summary

Prediction markets are markets for contracts that yield payments based on the outcome of an uncertain future event, such as a presidential election. Using these markets as forecasting tools could substantially improve decision making in the private and public sectors. We argue that U.S. regulators should lower barriers to the creation and design of prediction markets by creating a safe harbor for certain types of small stakes markets. We believe our proposed change has the potential to stimulate innovation in the design and use of prediction markets throughout the economy, and in the process to provide information that will benefit the private sector and government alike.

Introduction

Prediction markets are markets for contracts that yield payments based on the outcome of an uncertain future event, such as a presidential election, the release date for new software, or the action taken by the Federal Reserve on short-term interest rates. A key benefit is that the market price of these contracts can potentially provide more accurate forecasts of future events than other methods. Using these markets as forecasting tools could substantially improve decision making in the private and public sectors. They also can help manage risk more efficiently. It is precisely because prediction markets have great potential that we think the government should facilitate rather than hinder the introduction of these markets.

There are significant regulatory barriers to establishing prediction markets in the United States, in part because they are potentially subject to gambling laws. We argue that U.S. regulators should lower barriers to the creation and design of prediction markets by creating a safe harbor for certain types of small stakes markets. We believe our proposed change has the potential to stimulate innovation in the design and use of prediction markets throughout the economy, and in the process to provide information that will benefit the private sector and government alike.

[…]

Conclusion

We believe prediction markets can significantly improve decision making in both the private and public sectors. One of the clear benefits of allowing small stakes, non-profit markets to operate would be the greater use of prediction markets to inform both public and private decision making. A second benefit would be that access to better information could promote greater transparency and accountability in decision making. A third benefit might be that other countries and regions would promote prediction markets with more sensible regulation. Finally, we think there would be benefits from the development of new knowledge on how to design prediction markets.

We are aware that Congress did not intend the CFTC to regulate gambling and we believe that it is important to design this safe harbor in such a fashion that socially valuable prediction markets can get in, but gambling markets cannot.

Prediction markets have great potential for improving economic welfare and the decisions of private and public institutions alike. To help achieve that potential, the regulatory impediments to the use of prediction markets in the U.S. should be lowered. Here, we have suggested one approach for reducing those regulatory barriers.

AEI-Brookings Joint Center – The views in this paper represent those of the authors and do not necessarily represent the views of the institutions with which they are affiliated.

Kenneth J. Arrow – Stanford University

Robert Forsythe – University of South Florida

Michael Gorham – Illinois Institute of Technology

Robert Hahn – AEI-Brookings Joint Center

Robin Hanson – George Mason University

Daniel Kahneman – Princeton University

John O. Ledyard – California Institute of Technology

Saul Levmore – University of Chicago

Robert Litan – AEI-Brookings Joint Center

Paul Milgrom – Stanford University

Forrest D. Nelson – University of Iowa

George R. Neumann – University of Iowa

Charles R. Plott – California Institute of Technology

Thomas C. Schelling – University of Maryland

Robert J. Shiller – Yale University

Vernon L. Smith – George Mason University

Erik Snowberg – Stanford University

Cass R. Sunstein – University of Chicago

Paul C. Tetlock – University of Texas at Austin

Philip E. Tetlock – University of California at Berkeley

Hal R. Varian – University of California at Berkeley

Marco Ottaviani – London Business School

Justin Wolfers – University of Pennsylvania

Eric Zitzewitz – Stanford University

&#8211-

Previous: Statement on Prediction Markets – by Robert Hahn – 2007-05-07

This is Part II of a series of posts examining alternatives to the Continuous Double Auction market. Part I, posted at my CASTrader Blog, examines the problems with Continuous Double Auctions. This Part II examines some existing alternatives to them.

Invention of new types of markets has experienced a relative renaissance in recent years (mainly by the people who hang around Midas Oracle), many of which I surveyed before on my blog, and new types of markets are being invented as we speak (PDF). It&#8217-s hard to keep up, and I will caution that I am by no means any kind of expert on market design. That said, let&#8217-s examine an old-school market, as well as one of the new inventions.

Call Auction Market. Ironically, what the New York Stock Exchange replaced in the 1800s when they adopted CDAs has some interesting properties and advantages. A call market is typically organized as a price scan auction which basically amounts to this: poll every market participant and ask them how much they would tenatively offer to buy or sell at a given price. The search continues until buy/sell demand is balanced, at which price the market is cleared. The call market was rightly abandoned in the 1800s as markets grew due to the impossibilities of managing them in the pre-electronic era. I imagine it was mind-numbingly boring for traders as well. Recently, though, other types of call markets, such as crossing networks that batch orders at prices set in CDA markets have re-emerged, and some researchers have called for a major revival of the old call market (you may want to turn your sound off before clicking that link). Call markets have the interesting property of higher liquidity and lower short-term volatility relative to a CDA. When you realize CDAs are a sequential operation, and call markets are batch, it&#8217-s easy to see why this is true. In a batch operation, buy and sell orders are likely to offset, keeping price movement to a minimum vs. a bunch of sequential order fills alternating at the bid/ask. The advantage shifts towards price-takers, resulting in more trading and better price discovery, in my opinion. It&#8217-s not hard to see that a call auction where everyone&#8217-s offer was secret (and treated equally) would eliminate many of the shenanigans of CDAs. The following was said of call auction markets:

“Recent advances in computer technology have considerably expanded the call auction&#8217-s functionality. We suggest that the problems we are facing concerning liquidity, volatility, fragmentation and price discovery are largely endemic to the continuous market, and that the introduction of electronic call auction trading in the U.S. would be the most important innovation in market structure that could be made.”

That was said over 10 years ago, and it&#8217-s not hard to see why call markets are attractive, so why aren&#8217-t they taking over? Aside from call market proponent&#8217-s conspiracy theories that the bad boys of Wall Street would lose out, there is the major problem of immediacy. You just don&#8217-t trade a call market whenever you want to. The now defunct Arizona Stock Exchange was a call market that cleared once a day, for example. While some might argue that clearing once a day is a more productive use of people&#8217-s time, so long as some other market is clearing the same securities continuously, trading will flow to the other market, because there are simply more opportunities there.

Hanson&#8217-s Combinatorial Market. I&#8217-ve been fascinated by Hanson&#8217-s combinatorial market, although I must admit I don&#8217-t fully understand all of it&#8217-s intricacies. This market doesn&#8217-t use a limit order book at all (unless you want to use one because you want it to scale well), liquidity is always present via a market maker, and there is no bid-ask spread, because the price you pay is a continuous function of how much you want to buy or sell. The less the amount, the less your price will diverge from the last trade price. In the absence of other friction, the smallest trades are possible, even efficient, because liquidity is continuous (via a market maker that has a continuous price function). What&#8217-s more, you can have a functioning market with just a few traders, unlike a CDA. Hanson&#8217-s market is designed to function well in thin markets. Unfortunately, all of these characteristics are provided by one market maker adapted more for event markets than securities markets. This market maker decides what the price will be rather than the market players themselves. Furthermore, the market maker can be set up with different behaviors (scoring rules) and is subject to losing money (although the bounds of the loss is known ahead of time). While Hanson&#8217-s market maker may be an ideal way to subsidize liquidity in a fledgling prediction market, it doesn&#8217-t appear to me to be adaptable to a securities market.

The ideal market. An ideal dark market for CASTrader would be one that operated continuously, scaled well, and is general purpose like a CDA, while having the efficiency, volatility and trade encouraging characteristics of a call market, combined with the continuous liquidity and ability to function in thin markets of Hanson&#8217-s combinatorial market. To boot, I&#8217-d like it to be fair to traders of all sizes, as well as easy to program relative to a CDA. Is that possible? See Part III, over at CASTrader.

by Google&#8217-s Bo Cowgill (Project Manager):

&#8212-

At Google, we&#8217-re constantly trying to find new ways to organize the world&#8217-s information, including information relevant to our business. Building on the ideas of Friedrich Hayek and the Iowa Electronic Markets, a few Googlers (Doug Banks, Patri Friedman, Ilya Kirnos, Piaw Na and me, with some help from Hal Varian), set up a [prediction] market system inside the company.

The markets were designed to forecast product launch dates, new office openings, and many other things of strategic importance to Google. So far, more than a thousand Googlers have bid on 146 events in 43 different subject areas (no payment is required to play).

We designed the market so that the price of an event should, in theory, reflect a consensus probability that the event will occur. To determine accuracy of the market, we looked at the connection between prices of events and the frequency with which they actually occurred. If prices are correct, events priced at 10 cents should occur about 10 percent of the time.

In the graph below, the X-axis indicates the price ranges for the group. The orange line represents the average price, which is how often outcomes in that group should actually happen according to market prices. The purple line is how often they did happen. Ideally these would be equal, and as you can see they&#8217-re pretty close. So our prices really do represent probabilities – very exciting!

We also found that the market prices gave decisive, informative predictions in the sense that their predictive power increased as time passed and uncertainty was resolved. When a market first opens there may be considerable uncertainty about what will eventually happen- but as time goes on, some outcomes became more likely than others. The market prices should reflect this phenomenon, with the implied probability distributions becoming more concentrated over time.

Being geeks, we naturally used information theory to measure the entropy of our probability distributions:

In this graph, we have weeks before market expiration on the X-axis, and entropy (in bits) on the Y-axis. We&#8217-ve included some reference entropies to help your intuition, and you can see that in addition to accurate predictions, the distributions become steadily more informative and decisive (lower entropy) over time.

Our search engine works well because it aggregates information dispersed across the web, and our internal [prediction] markets are based on the same principle: Googlers from across the company contribute knowledge and opinions which are aggregated into a forecast by the market. Sometimes, just feeling lucky isn&#8217-t enough, and these tools can help.

&#8212-

Putting crowd wisdom to work – by Bo Cowgill – 2005-09-21 – Published originally on the Official Google Blog