Predicting the future through game theory

[nmap]

This is an odd one - but, perhaps, worthy of consideration.

It appears that a professor of political science has found a way, using computers and game theory, to accurately predict world events. He has, according to third-party reports, established a favorable track record with the CIA.

The first link discusses him, his background, and his methods.

The second link is a 20 minute video with predictions for Iran and its nuclear program.

LINK 1

LINK 2

Sadly, there are no stock market predictions.

[Kyobanim]

I saw that guy on the Science Channel a while back. His predictions were pretty accurate. Very interesting stuff.

[MILON]

I just saw Mr. Bueno de Mesquita on the History channel tonight actually. The show was comparing him to Nostradamus. Interesting show.

http://www.history.com/shows.do?acti...isodeId=428212

[Go Devil]

Thank you NMAP,


What I found most interesting in Mr. Bueno de Mesquitas talk is stressed at 16:47 in the link that NMAP provided.

At 17:50 in the link, I looked in the mirror and noticed that my tinfoil hat had increased in size.

I would like to see their predictions on the American populace.

[GreenSalsa]

Thanks for the link--I will review it a little later.

Game theory is one of the tools they use to teach us here, at Naval Postgraduate School. In fact, they brought John Nash (man behind the movie "A Beautiful Mind") here to talk to us. I found the subject interesting and nearly wrote my thesis on it.

In essence, it takes seemingly "fuzzy" choices and quantifies them in such a manner that makes it possible to determine if it is best to move first, second, make threats, promises, or not play at all.

Really interesting stuff.

[Warrior-Mentor]

Thanks for the POSTS!

I considered going to TED this year...now I'm kicking myself for not going....

[CSB]

I use a form of Game Theory in my legal practice as a criminal defense attorney. It halps me work a case and give my client meaningful advice on whether or not to seek a plea agreement, and if so, what the terms of that plea agreement should be. Consider this situation:

FACTS:

18 year old kid is getaway driver for a armed robbery at a Kentucky Fried Chicken. He’s caught shortly thereafter and the robber jumps from the car and runs. The Defendant admits to being the driver, but denies knowing it was an armed robbery, says the robber was just going to grab the money and run when the register was open.

He is indicted for Aggravated (Armed) Robbery, even though he didn’t have the gun, under Tennessee’s law tersely entitled “criminal responsibility for the acts of another.”

He wants to deal the case. What’s fair?

My analysis:

Chance of an absolute acquittal, slim, but never zero. Maybe 10% since he is young, has a clean record, cooperated, and didn’t have the gun.

Chance of being convicted of “Facilitation of Armed Robbery,” a lesser included offense that carries 3 ~ 6 years, about 50%. Of that, about a half and half chance that the judge will give him the minimum of 3 years (if he comes across as young and dumb at trial/sentencing); and an equal chance that the judge will give him the maximum six years (if he comes across as a hard head who knew that the robber was probably armed).

Chance of being convicted of the indicted charge “Aggravated (Armed) Robbery by criminal responsibility for the acts of another,” oh, about 25%. That carries a range of punishment from a minimum of 8 years, to a maximum of 12 years, which I apportion 4 out of 5 chance of getting the minimum, but a one out of five chance of getting the maximum.

SO WHAT’s THE “VALUE OF THE GAME.” Damned serious matters, but in Game Theory it is still called “the value of the game.” Chance of conviction = 90%, so a plea agreement is clearly in his best interest. And the terms?

The math:

Acquittal - 10% x zero = 0
Facilitation - 25% x 3 years = .75 years
Facilitation - 25% x 6 years = 1.5 years
Aggravated Robbery - 20% x 8 years = 1.6 years
Aggravated Robbery - 5% x 12 years = .6 years

TOTAL: = 3.325 years

Those 3.325 years represent a weighted average. It’s my best analysis that if I were to try 100 of these cases, on average defendant would walk out with a sentence of 3 1/3 years to serve.

So if the District Attorney offers to settle for a conviction of “Facilitation, three years to serve,” take it, it’s better than the value of the game.

If the District Attorney offers “Aggravated Robbery for the minimum 8 years,” reject it, that exceeds the value of the game.

Sometimes the District Attorney and I will actually sit down and argue the percentages, and run the figures several times before reaching an agreement. I’m sure the judge wonders sometime why a case is settled for “Plea of Guilty of Facilitation of Aggravated Robbery, Sentence of 3 years and 4 Months.”

[The Reaper]

That is going to be a life altering experience for the young man, about to become a convicted felon.

Any chance of pleading down to a misdemeanor and no jail time for rolling over on the robber?

On the OT, I think that the game theory would require rational actions from players, and NK, Iran, Venezuela, etc. are not rational and in fact, may select CoAs not in their own best interest. I find that they continue this behavior, even after we whacked Saddam twice for that, and both times, he should have seen it coming.

TR

[GratefulCitizen]

Originally Posted by The Reaper

On the OT, I think that the game theory would require rational actions from players, and NK, Iran, Venezuela, etc. are not rational and in fact, may select CoAs not in their own best interest. I find that they continue this behavior, even after we whacked Saddam twice for that, and both times, he should have seen it coming.

TR

This is the crux of the matter.

MAD with the USA vs. USSR was a classic case of a Nash equilibrium.
Both parties had rational self-interest.

When a culture finds glory in a lose-lose outcome (win-lose in their eyes), dire solutions may be necessary.
Harry Truman understood this.

[Slantwire]

Originally Posted by GratefulCitizen

When a culture finds glory in a lose-lose outcome (win-lose in their eyes), dire solutions may be necessary.
Harry Truman understood this.

I believe part of the system tries to account for that. They ARE rational, within their own outlook. WW2 Japan placed such a powerful negative emphasis on surrender that, within that framework, fighting to the death was the rational thing to do.

The crazies always think that they're the sane ones.

[Richard]

If they decide to call this thing the WOPR* and it wants to begin playing Global Thermo-Nuclear War...better switch it to Tic-Tac-Toe or we'll be toast!

"Shall we play a game?"

Richard's $.02

* http://www.youtube.com/watch?v=tAcEzhQ7oqA

[Hammock]

Here are links to two podcasts from Econtalk with Prof. de Mesquita.

In this one, he talks about the incentives facing dictators and democratic leaders. Both have to face competition from rivals. Both try to please their constituents and cronies to stay in power. He applies his insights to foreign aid, the Middle East, Venezuela, the potential for China's evolution to a more democratic system, and Cuba. Along the way, he explains why true democracy is more than just elections--it depends crucially on freedom of assembly and freedom of the press.

http://www.econtalk.org/archives/200..._bueno_de.html

In the second one, he talks to EconTalk host Russ Roberts about threats to U.S. security, particularly Iran. Bueno de Mesquita argues that Iran is of little danger to the United States and that Ahmadinejad is an unimportant player in Iran's political system, more of a stalking horse for provocative ideas rather than a wielder of power. Bueno de Mesquita then looks at what Iran has to gain and to lose by appearing to build a nuclear weapons program and actually using a nuclear weapon. He then goes on to examine the nature of other threats to the United States. The closing topic of the conversation is the peculiar incentives facing U.S. Presidents as their terms expire.

http://www.econtalk.org/archives/200..._mesqui_1.html

You can download or play directly at the links.

[CSB]

That is going to be a life altering experience for the young man, about to become a convicted felon.

Any chance of pleading down to a misdemeanor and no jail time for rolling over on the robber?

On the OT, I think that the game theory would require rational actions from players, and NK, Iran, Venezuela, etc. are not rational and in fact, may select CoAs not in their own best interest. I find that they continue this behavior, even after we whacked Saddam twice for that, and both times, he should have seen it coming.

Right on about the "misdemeanor" vs. "felony" factor.

Reaper pointed out two major weaknesses in Game Theory when applied to interpersonal relationships:

1 - The same "payoff" (what you win or lose) is not necessarily of equal value to both players. For me, losing a $2,000 bet would be a major hit to my monthly finances. To Bill Gates, losing a $2,000 bet is but petty cash.

2 - The goal for one party is not necessarily the "anti-goal" of the opposing party.

I recently wrote a brief paper for my daughter as part of her history class. It included the following:

Consider the first Gulf War (Desert Shield/Desert Storm).

Make a list of the objectives of the allied Forces,
and the objectives of Saddam Hussian:

ALLIES
1 - Liberate Kuwait.
2 – Minimize Allied Casualties.
3 – Encourage Democracy in Iraq.
4 – Discourage Iran from aggression.

SADDAM HUSSIAN
1 – Stay Alive.
2 – Stay in Power.
3 – Steal from Kuwait.
4 - Humilate Israel, as leader of Muslims.


How did each side do?

The first priority of a dictator is to stay alive. (Ask Romanian Dictator Nicolas Ceausescu, Emperor Louis XVI of France, or Italy’s Benito Musolini). Second best is to stay in power and stay alive, but the first objective always outweighs the second objective. (Ask Ferdinand Marcos, Idi Amin, the Shah of Iran, etc.). It is better for a dictator to cut and run, rather than stay and be cut.

And notice that there was no corresponding opposite objective on the part of the Allies: “Kill Saddam Hussian” or even “remove Saddam Hussian from power.”

On the Saddam Hussian side, there is no corrresponding “keep control of Kuwait” "anti-goal" objective to the Allied objective of “Liberate Kuwait.”

By the end of the first Gulf War, Allied forces had accomplished Objectives 1 and to a large extent, Objective 2. Objective 3 was a failure, and no one knows is Iran has withheld aggressive moves because of Allied intervention or becaue it was/is busy attempting to manufacture nuclear weapons.

As for Saddam Hussian, he accomplished all of his objectives. In particular, the SCUD missile attacks on Israel, with Israel making no reponse against Iraq, boosted his image in the Muslim world.

Some commentators asked, at the end of the first Gulf War, why Saddam Hussian didn’t surrender. Didn’t he know he had been beat, they asked?

The answer is simple: No, he didn’t know he had been beat. In his eyes, he had actually won. He was alive, in power, and the emerging leader of the Muslim world.

Game theory falls apart under such conditions.

[redleg99]

I think that the game theory would require rational actions from players, and NK, Iran, Venezuela, etc. are not rational and in fact, may select CoAs not in their own best interest.

I believe part of the system tries to account for that. They ARE rational, within their own outlook.

Reaper pointed out two major weaknesses in Game Theory when applied to interpersonal relationships:

1 - The same "payoff" (what you win or lose) is not necessarily of equal value to both players. For me, losing a $2,000 bet would be a major hit to my monthly finances. To Bill Gates, losing a $2,000 bet is but petty cash.

2 - The goal for one party is not necessarily the "anti-goal" of the opposing party.

I’m still quite new at game theory, but I think what you are all describing is called “choice preference” in game theory terminology.

In other words, the accounting of different preferences among the players.

Although I didn’t hear him actually say the words “choice preference,” this seems to be one of the things Professor Bueno de Mesquita discusses between 8:11 and 11:10 on the video.

[Guy]

Unless you're playing it for REAL! Then it's no longer a game and/or theory...

Stay safe.