This article is rated B-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | |||||||||||
|
According to Dr. Martin Shubik the formal precommitment mechanism in 1971, needed to know that there was atleast 2 or more precommitted bidders in the auction event before the decision to initiate the bidding phase was made.
In 2012, the formal precommitment mechanism was converted into a computer algorithm, also known as the prefunding algorithm.
Dr. Martin Shubik has formally admitted that prefunding algorithm can be considered as the original way of doing business in 1971.
It was considered commonsense knowledge in 1971 that auction needed to have atleast 2 or more precommitted bidders before the decision to initiate an auction was made.
Bids were sold under "use them or lose them" contract.
A dollar auction does not induce rational players to overpay for the dollar. It is trivially easy to understand this. All players have the option to earn $0 by never bidding. Thus, each player's expected payoff must be at least 0. Further, the game has a clean solution found here: http://www.math.toronto.edu/mpugh/Teaching/Sci199_03/dollar_auction_2.pdf
The article needs to be rewritten and framed using the above article. As is, the article is wrong. — Preceding unsigned comment added by 75.84.163.162 ( talk) 02:40, 24 December 2014 (UTC)
There is a "2 or more precommitted bidders" auction initiation criteria that has been ignored for quite some time. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 12:06, 30 October 2016 (UTC)
I've expanded the Setup section, as without my additions, the description was insufficiently detailed and it did not explain clearly enough what happens to anyone unfamiliar with the topic. I used my own lecture notes which were based on the Colman reference as a guide, and added a reference to that book in an appropriate place later on. I hope that doesn't upset anyone here. DDStretch (talk) 12:44, 27 July 2007 (UTC)
I am a professional mathematician but I do not understand this article at all...Why would the initial 1 cent bidder lose his 1 cent?? The rules said that only the second highest bidder would pay. Now there are three bids (1, 2, and 3 cents) from three different bidders. Then the first guy should not be obliged to pay?????
Isn’t it a winning strategy not to play? Roman V. Odaisky 20:40, 28 July 2007 (UTC)
"Additionally, this game is not an auction, per say, but gambling. As gambling, the choice to enter into it with more money than one is willing to lose is an irrational decision. The claim is perfect information, but calling gambling an auction confuses people and distorts one's perception of how to calculate returns, resulting in actions consistent with auctions prior to the $1.00 bid, and actions consistent with gambling after that point." contributed by 70.152.255.102. Pete.Hurd 15:25, 24 September 2007 (UTC)
If this was "re"moved, why, or if it was simply moved, where can I find it? --the person who added this notation. —Preceding unsigned comment added by 132.170.49.6 ( talk) 19:07, 11 October 2007 (UTC)
All auctions, on some level, relate to gambling: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=905606 This auction is more closely related to gambling because of the additional requirement that the losing bidder must also pay: http://en.wikipedia.org/wiki/Betting_(poker) VS http://en.wikipedia.org/wiki/Auction It could be considered a type of "all-pay" auction, which is generally used in charity cases, where all bidders are not in it so much for the prize as much as wanting to help the charity. So yes, it is an auction, but not one in the traditional sense. Tone: If your claim is that my writing had a tone (which was certainly not my intent) then how can you argue that using a certain 'tone' of words in describing the game would not change anything? Perfect Information and Gambling? http://links.jstor.org/sici?sici=0363-7425(198604)11%3A2%3C311%3AECTACO%3E2.0.CO%3B2-U Escalation of Commitment is extremely rare in cases of Perfect Information. The most obvious case where it does occur, is gambling. That was what I was trying to explain.20:09, 19 November 2007 (UTC)
Just out of curiosity, why is there a picture of a dollar in here?
The dollar's not in the act of being auctioned, most people who read this article are going to know what a dollar looks like, and in any case, dollar auctions (or, I suppose, currency auctions) can just as easily use any currency.
Anybody care to defend the image? superluser t c 2007 December 23, 15:30 (UTC)
Hello,
Though personally I don't really care that much and the pictureless article is fine to me, but I must note that I'm a visual thinker and an illustrative addition does help me in my mental processes. I just came here from an article about this sunject in another language where a dollar picture was present, and I've found myself to focus on the image while organizing and musing about the procedures in my head, so it basically served as an external anchor. Since it is a very simple subject I didn't really need this anchor, all the visualising is done is my head, but in more complex articles- pictures are a must. Shame that the wiki format can't contain videos, since the easiest way to learn for me are those videos you see on History/Discovery channel...
The first paragraph of the "Refutations" section should be removed. The sort of collaborations it describes violate the premise of a game in game theory. —Preceding unsigned comment added by Emperor Will ( talk • contribs) 21:09, 26 August 2008 (UTC)
I think the equilibrium reversions are kinda missing the point. First, to the editor who's saying there is a unique equilibrium--unfortunately, offering your own proof probably isn't good enough for Wikipedia. WP:OR generally means you need to cite someone else's proof. (I personally would be interested in seeing your proof on my talk page, if you can't point toward one in the literature.) On the other hand, the "no equilibrium" bit that's there isn't cited either right now, so I don't know that we should be very enthusiastic to revert your change. I'm hoping to look this stuff up later today and see if I can flesh it out. CRETOG8( t/ c) 17:34, 10 September 2008 (UTC)
The equilibrium that exists is the same as balance of power between global powers and the world order.
The ones in power will force the players to a contract, ie. Power Elites forced European leaders to form EU under contract or Europe would face another war.
The power elites would start the war unless all EU country leaders agree to contract terms unitedly.
In other words the citizens and players were forced to agree to a united contract since the "bids" were sold in advance and "bids" were sold on use them, or lose them terms. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 09:36, 30 October 2016 (UTC)
I have two questions about the current presentation of the game in the article (and in general). First, Shubik presented the game as having a $0.05 bid increment. I would lean toward following him, but I haven't read the other sources yet, and $0.01 increments work to. Thoughts?
Second, I'm a little stumped at the moment about the tie-breaking rules, which should be detailed for the game to be complete. Shubik's a little confusing on this point (to me). CRETOG8( t/ c) 22:13, 10 September 2008 (UTC)
I don't think it should be linked here... Sure, the article explains that the dollar auction is irrational, but linking to the site still seems a lot like free advertising to me (especially since the site's relevance isn't explained, and since it isn't exactly the same game). sten for the win ( talk) 00:56, 17 December 2008 (UTC)
Swoopo was mentioned here again, with a WARNING--There's a lot of bloggish reference to Swoopo and the dollar auction, but I haven't found a ref good enough to use in the article. I think a mention of it might be OK if well-referenced. CRETOG8( t/ c) 00:34, 15 September 2009 (UTC)
There's a bit from Richard Thaler in the NYT which makes the connection between the dollar auction and Swoopo.com. It's here. I think that's a good enough ref if someone wants to include it. CRETOG8( t/ c) 04:00, 15 November 2009 (UTC)
Some recent edits removed "fact" tags on some of the stuff in the "Refutations" section. Whether the stuff is obvious or not, the thing is that WP requires verifiability not truth. The important thing for this article is that we aren't coming up with our own refutations, but describing possible refutations that others have come up with. CRETOG8( t/ c) 18:28, 19 January 2009 (UTC)
An encyclopedia aims to document externally existing knowledge of a topic. A great deal has been written about the dollar auction, and that body of work comprises the subject of the encyclopedia article. Thought outside of that work is appropriate in other venues, but not appropriate for this venue. The prescription against reporting original thought WP:OR, or representing views out of proportion to their prominence in reliable sources WP:UNDUEWEIGHT, means that we only present that which is typically presented on a topic by secondary and tertiary sources, see WP:SECONDARY. Pete.Hurd ( talk) 05:42, 21 January 2009 (UTC)
I removed the reference to the expected value. The concept of expected value makes sense only in context of random variables. As there are no clearly defined random variables here, the reference to the expected value does not make sense. I assume this part of the article was original research as no citation was also provided.
seriously, im sure scholars and internet people can debate this to death and it has its implications for the definition of 'rational player' and such.... but is it that hard to find a 'generous benefactor' to put $1 up for auction. not as an undisputed answer to this paradox, that would be like experimentally testing Schrödinger cat. it would be a nice demonstration about how rational people can diverge from the 'perfectly rational' model and get a better outcome. just like the travelers dilemma. I predict that smart players would end up at a tie bid of either 50 cents or 1 dollar. at 50 cents they break even. if they dont have the foresight to stop at 50 cents then surely they would stop at 1 dollar because at that point it is no longer profitable to either player. —Preceding unsigned comment added by 96.234.107.175 ( talk) 15:40, 10 July 2009 (UTC)
out 88% of the 10,000 weon note (roughly worth $10). Kdammers ( talk) 07:03, 5 November 2013 (UTC)
See my question here on the "Polish Auction" which is sometimes known as a "Dollar-Dollar Auction". While I doubt these are directly related to the main subject of this article I was wondering if anyone had run into any game theory research on the subject. I figure the similarity in names might have caused a serendipitous connection. :) 66.97.213.94 ( talk) 19:49, 11 October 2010 (UTC)
I removed the stuff about the cooperative outcome from the Refutations section as 1) it's OR, 2) it's not really a refutation since it involves a fundamentally different game, 3) is not even particularly insightful since with cooperative behavior essentially anything can happen.
What I left in the section isn't particularly accurate either however. As mentioned above "don't play in the first place" is not an equilibrium either. If everyone else chooses "bid zero" then your best option is to "bid 1 cent" (or epsilon or whatever) since in that case you make 99 cent profit, as opposed to 0 profit if you also play "bid zero". Of course that means that someone else's best response is "bid 2 cents" and then the whole process begins. As I've stated above, this is essentially a game with no equilibrium (either in mixed or pure strategies) and that's what creates the paradox. Volunteer Marek ( talk) 05:40, 7 July 2011 (UTC)
The article claims that the "Dollar Auction" contains a paradox, but it is not clear where the paradox lies. What does "complete information" mean in this context? Can the bidder assume that the other bidders (if any) are rational? Consider these cases:
1. There is only one bidder, Alice. She knows there are no other bidders. She bids 5c. She wins. Done.
The formal precommitment mechanism made sure there is atleast 2 or more precommitted bidders in the room before auction was initiated.
2. There are other bidders, but Alice gets to bid first (and she knows that). She bids 95c. The other bidders are rational and nobody bids 1$ because bidding is extra work for no gain. Alice wins. Done.
The formal precommitment mechanism made sure there is atleast 2 or more precommitted bidders in the room before auction was initiated.
So, again, where is the paradox?
So there was atleast 2 persons who had bought bids towards the auction.
Bids were sold on "use them or lose them" contract.
Luigi Semenzato — Preceding
unsigned comment added by
2620:0:1000:3803:A800:1FF:FE00:54E8 (
talk)
01:12, 1 June 2013 (UTC)
As far as research goes, there is an academic article, but with the pay-wall, I can't say if it actually includes empirical research in spite of the statement about results on the first page: http://www.jstor.org/discover/10.2307/173254?uid=3738392&uid=2&uid=4&sid=21102869324217 . The author "ignores" both co-operation and the 99 cent solution. Another article >is< a report of an actual experiment. the fact that it was done with MBA students might explain the extreme results: http://jme.sagepub.com/content/26/1/56.abstract . Kdammers ( talk) 07:18, 5 November 2013 (UTC)
If the minimum bid is 5 cents how can the superrational guys bid 1 cent? Palosirkka ( talk) 06:08, 30 April 2014 (UTC)
According to Dr. Martin Shubik the formal precommitment mechanism in 1971, needed to know that there was atleast 2 or more precommitted bidders in the auction event before the decision to initiate the bidding phase was made.
In 2012, the formal precommitment mechanism was converted into a computer algorithm, also known as the prefunding algorithm.
Dr. Martin Shubik has formally admitted that prefunding algorithm can be considered as the original way of doing business in 1971.
It was considered commonsense knowledge in 1971 that auction needed to have atleast 2 or more precommitted bidders before the decision to initiate an auction was made. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 08:38, 30 October 2016 (UTC)
This article is rated B-class on Wikipedia's
content assessment scale. It is of interest to the following WikiProjects: | |||||||||||
|
According to Dr. Martin Shubik the formal precommitment mechanism in 1971, needed to know that there was atleast 2 or more precommitted bidders in the auction event before the decision to initiate the bidding phase was made.
In 2012, the formal precommitment mechanism was converted into a computer algorithm, also known as the prefunding algorithm.
Dr. Martin Shubik has formally admitted that prefunding algorithm can be considered as the original way of doing business in 1971.
It was considered commonsense knowledge in 1971 that auction needed to have atleast 2 or more precommitted bidders before the decision to initiate an auction was made.
Bids were sold under "use them or lose them" contract.
A dollar auction does not induce rational players to overpay for the dollar. It is trivially easy to understand this. All players have the option to earn $0 by never bidding. Thus, each player's expected payoff must be at least 0. Further, the game has a clean solution found here: http://www.math.toronto.edu/mpugh/Teaching/Sci199_03/dollar_auction_2.pdf
The article needs to be rewritten and framed using the above article. As is, the article is wrong. — Preceding unsigned comment added by 75.84.163.162 ( talk) 02:40, 24 December 2014 (UTC)
There is a "2 or more precommitted bidders" auction initiation criteria that has been ignored for quite some time. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 12:06, 30 October 2016 (UTC)
I've expanded the Setup section, as without my additions, the description was insufficiently detailed and it did not explain clearly enough what happens to anyone unfamiliar with the topic. I used my own lecture notes which were based on the Colman reference as a guide, and added a reference to that book in an appropriate place later on. I hope that doesn't upset anyone here. DDStretch (talk) 12:44, 27 July 2007 (UTC)
I am a professional mathematician but I do not understand this article at all...Why would the initial 1 cent bidder lose his 1 cent?? The rules said that only the second highest bidder would pay. Now there are three bids (1, 2, and 3 cents) from three different bidders. Then the first guy should not be obliged to pay?????
Isn’t it a winning strategy not to play? Roman V. Odaisky 20:40, 28 July 2007 (UTC)
"Additionally, this game is not an auction, per say, but gambling. As gambling, the choice to enter into it with more money than one is willing to lose is an irrational decision. The claim is perfect information, but calling gambling an auction confuses people and distorts one's perception of how to calculate returns, resulting in actions consistent with auctions prior to the $1.00 bid, and actions consistent with gambling after that point." contributed by 70.152.255.102. Pete.Hurd 15:25, 24 September 2007 (UTC)
If this was "re"moved, why, or if it was simply moved, where can I find it? --the person who added this notation. —Preceding unsigned comment added by 132.170.49.6 ( talk) 19:07, 11 October 2007 (UTC)
All auctions, on some level, relate to gambling: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=905606 This auction is more closely related to gambling because of the additional requirement that the losing bidder must also pay: http://en.wikipedia.org/wiki/Betting_(poker) VS http://en.wikipedia.org/wiki/Auction It could be considered a type of "all-pay" auction, which is generally used in charity cases, where all bidders are not in it so much for the prize as much as wanting to help the charity. So yes, it is an auction, but not one in the traditional sense. Tone: If your claim is that my writing had a tone (which was certainly not my intent) then how can you argue that using a certain 'tone' of words in describing the game would not change anything? Perfect Information and Gambling? http://links.jstor.org/sici?sici=0363-7425(198604)11%3A2%3C311%3AECTACO%3E2.0.CO%3B2-U Escalation of Commitment is extremely rare in cases of Perfect Information. The most obvious case where it does occur, is gambling. That was what I was trying to explain.20:09, 19 November 2007 (UTC)
Just out of curiosity, why is there a picture of a dollar in here?
The dollar's not in the act of being auctioned, most people who read this article are going to know what a dollar looks like, and in any case, dollar auctions (or, I suppose, currency auctions) can just as easily use any currency.
Anybody care to defend the image? superluser t c 2007 December 23, 15:30 (UTC)
Hello,
Though personally I don't really care that much and the pictureless article is fine to me, but I must note that I'm a visual thinker and an illustrative addition does help me in my mental processes. I just came here from an article about this sunject in another language where a dollar picture was present, and I've found myself to focus on the image while organizing and musing about the procedures in my head, so it basically served as an external anchor. Since it is a very simple subject I didn't really need this anchor, all the visualising is done is my head, but in more complex articles- pictures are a must. Shame that the wiki format can't contain videos, since the easiest way to learn for me are those videos you see on History/Discovery channel...
The first paragraph of the "Refutations" section should be removed. The sort of collaborations it describes violate the premise of a game in game theory. —Preceding unsigned comment added by Emperor Will ( talk • contribs) 21:09, 26 August 2008 (UTC)
I think the equilibrium reversions are kinda missing the point. First, to the editor who's saying there is a unique equilibrium--unfortunately, offering your own proof probably isn't good enough for Wikipedia. WP:OR generally means you need to cite someone else's proof. (I personally would be interested in seeing your proof on my talk page, if you can't point toward one in the literature.) On the other hand, the "no equilibrium" bit that's there isn't cited either right now, so I don't know that we should be very enthusiastic to revert your change. I'm hoping to look this stuff up later today and see if I can flesh it out. CRETOG8( t/ c) 17:34, 10 September 2008 (UTC)
The equilibrium that exists is the same as balance of power between global powers and the world order.
The ones in power will force the players to a contract, ie. Power Elites forced European leaders to form EU under contract or Europe would face another war.
The power elites would start the war unless all EU country leaders agree to contract terms unitedly.
In other words the citizens and players were forced to agree to a united contract since the "bids" were sold in advance and "bids" were sold on use them, or lose them terms. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 09:36, 30 October 2016 (UTC)
I have two questions about the current presentation of the game in the article (and in general). First, Shubik presented the game as having a $0.05 bid increment. I would lean toward following him, but I haven't read the other sources yet, and $0.01 increments work to. Thoughts?
Second, I'm a little stumped at the moment about the tie-breaking rules, which should be detailed for the game to be complete. Shubik's a little confusing on this point (to me). CRETOG8( t/ c) 22:13, 10 September 2008 (UTC)
I don't think it should be linked here... Sure, the article explains that the dollar auction is irrational, but linking to the site still seems a lot like free advertising to me (especially since the site's relevance isn't explained, and since it isn't exactly the same game). sten for the win ( talk) 00:56, 17 December 2008 (UTC)
Swoopo was mentioned here again, with a WARNING--There's a lot of bloggish reference to Swoopo and the dollar auction, but I haven't found a ref good enough to use in the article. I think a mention of it might be OK if well-referenced. CRETOG8( t/ c) 00:34, 15 September 2009 (UTC)
There's a bit from Richard Thaler in the NYT which makes the connection between the dollar auction and Swoopo.com. It's here. I think that's a good enough ref if someone wants to include it. CRETOG8( t/ c) 04:00, 15 November 2009 (UTC)
Some recent edits removed "fact" tags on some of the stuff in the "Refutations" section. Whether the stuff is obvious or not, the thing is that WP requires verifiability not truth. The important thing for this article is that we aren't coming up with our own refutations, but describing possible refutations that others have come up with. CRETOG8( t/ c) 18:28, 19 January 2009 (UTC)
An encyclopedia aims to document externally existing knowledge of a topic. A great deal has been written about the dollar auction, and that body of work comprises the subject of the encyclopedia article. Thought outside of that work is appropriate in other venues, but not appropriate for this venue. The prescription against reporting original thought WP:OR, or representing views out of proportion to their prominence in reliable sources WP:UNDUEWEIGHT, means that we only present that which is typically presented on a topic by secondary and tertiary sources, see WP:SECONDARY. Pete.Hurd ( talk) 05:42, 21 January 2009 (UTC)
I removed the reference to the expected value. The concept of expected value makes sense only in context of random variables. As there are no clearly defined random variables here, the reference to the expected value does not make sense. I assume this part of the article was original research as no citation was also provided.
seriously, im sure scholars and internet people can debate this to death and it has its implications for the definition of 'rational player' and such.... but is it that hard to find a 'generous benefactor' to put $1 up for auction. not as an undisputed answer to this paradox, that would be like experimentally testing Schrödinger cat. it would be a nice demonstration about how rational people can diverge from the 'perfectly rational' model and get a better outcome. just like the travelers dilemma. I predict that smart players would end up at a tie bid of either 50 cents or 1 dollar. at 50 cents they break even. if they dont have the foresight to stop at 50 cents then surely they would stop at 1 dollar because at that point it is no longer profitable to either player. —Preceding unsigned comment added by 96.234.107.175 ( talk) 15:40, 10 July 2009 (UTC)
out 88% of the 10,000 weon note (roughly worth $10). Kdammers ( talk) 07:03, 5 November 2013 (UTC)
See my question here on the "Polish Auction" which is sometimes known as a "Dollar-Dollar Auction". While I doubt these are directly related to the main subject of this article I was wondering if anyone had run into any game theory research on the subject. I figure the similarity in names might have caused a serendipitous connection. :) 66.97.213.94 ( talk) 19:49, 11 October 2010 (UTC)
I removed the stuff about the cooperative outcome from the Refutations section as 1) it's OR, 2) it's not really a refutation since it involves a fundamentally different game, 3) is not even particularly insightful since with cooperative behavior essentially anything can happen.
What I left in the section isn't particularly accurate either however. As mentioned above "don't play in the first place" is not an equilibrium either. If everyone else chooses "bid zero" then your best option is to "bid 1 cent" (or epsilon or whatever) since in that case you make 99 cent profit, as opposed to 0 profit if you also play "bid zero". Of course that means that someone else's best response is "bid 2 cents" and then the whole process begins. As I've stated above, this is essentially a game with no equilibrium (either in mixed or pure strategies) and that's what creates the paradox. Volunteer Marek ( talk) 05:40, 7 July 2011 (UTC)
The article claims that the "Dollar Auction" contains a paradox, but it is not clear where the paradox lies. What does "complete information" mean in this context? Can the bidder assume that the other bidders (if any) are rational? Consider these cases:
1. There is only one bidder, Alice. She knows there are no other bidders. She bids 5c. She wins. Done.
The formal precommitment mechanism made sure there is atleast 2 or more precommitted bidders in the room before auction was initiated.
2. There are other bidders, but Alice gets to bid first (and she knows that). She bids 95c. The other bidders are rational and nobody bids 1$ because bidding is extra work for no gain. Alice wins. Done.
The formal precommitment mechanism made sure there is atleast 2 or more precommitted bidders in the room before auction was initiated.
So, again, where is the paradox?
So there was atleast 2 persons who had bought bids towards the auction.
Bids were sold on "use them or lose them" contract.
Luigi Semenzato — Preceding
unsigned comment added by
2620:0:1000:3803:A800:1FF:FE00:54E8 (
talk)
01:12, 1 June 2013 (UTC)
As far as research goes, there is an academic article, but with the pay-wall, I can't say if it actually includes empirical research in spite of the statement about results on the first page: http://www.jstor.org/discover/10.2307/173254?uid=3738392&uid=2&uid=4&sid=21102869324217 . The author "ignores" both co-operation and the 99 cent solution. Another article >is< a report of an actual experiment. the fact that it was done with MBA students might explain the extreme results: http://jme.sagepub.com/content/26/1/56.abstract . Kdammers ( talk) 07:18, 5 November 2013 (UTC)
If the minimum bid is 5 cents how can the superrational guys bid 1 cent? Palosirkka ( talk) 06:08, 30 April 2014 (UTC)
According to Dr. Martin Shubik the formal precommitment mechanism in 1971, needed to know that there was atleast 2 or more precommitted bidders in the auction event before the decision to initiate the bidding phase was made.
In 2012, the formal precommitment mechanism was converted into a computer algorithm, also known as the prefunding algorithm.
Dr. Martin Shubik has formally admitted that prefunding algorithm can be considered as the original way of doing business in 1971.
It was considered commonsense knowledge in 1971 that auction needed to have atleast 2 or more precommitted bidders before the decision to initiate an auction was made. — Preceding unsigned comment added by MikoFilppula ( talk • contribs) 08:38, 30 October 2016 (UTC)