langreiter.com plain, simple: comments-2006-01-02

LANGREITER.COM plain, simple

START

Montag, 2. Januar 2006

So schaut also 2006 aus.

Bei dieser Gelegenheit ein grosses Dankeschön an meine Silvester-Hosts S. und T. aus M., es war ein tolles Wochen- und Jahresende!

2006-01-02-01

Chris Hibbert: An introduction to basic [create Prediction Market] formats [ Ø UNICAST • ]

Guan Yang's comments.

2006-01-02-pred

2005 in links, 2006 in predictions, assembled by andi.

2006-01-02-tyi

slauti 6882 days AGO
well, we should care more about prediction markets as they have become so important in the real world but they take up an awful lot of brainpower if you do not want to focus on them alone. I've been knowing about this for a long time as a friend and me did an options and futures simulator for a german media house in about 199early named Brokerking. I knew then that it would flop allthough people had bought share trading simulators like rolls. we had to go in and correct the author's (to whose (half) popular book we were adding the simulator) logic quite a bit, because it just did not work. these models incl. the classic Black-Scholes are just one step too complex for the average listener. even now I had to read the introduction twice having been out of the topic for a long while. on the other hand gy shows that extensive pullout like the example in his comment is very nice, because examples that show single steps to develop a position are a lot easier to understand.

chris 6879 days AGO
(At the moment) I'm most interested in Prediction Markets for widely distributed decision making (more realistically "facilitating") to be used inside organizations; I dabbled a bit in theory & practice when working on an (unrelated) project for a certain Austrian betting company but haven't really thought much about them since.

I'm happy Guan pointed us to Chris Hibbert's blog, it's great reading, e.g. (duh, obvious) wrt/ Google's internal markets and the issues they had to contend with (insiders). Also: "The main problem with trying to use Prediction Markets to foresee trends is that you have to ask the right question."

slauti 6879 days AGO
ask the right question is the most important thing. If you don't it's like futures trading on wheat when you know beforehand that world wide weather is going to let prices rise. But if not you can still take clues from the deviatons in the expected trend.

Second thought: You really do need large markets or medium sized quite expert markets to get anything realistic. Remember all the past election exchange markets in Austria? Short term prediction in the super simple form of shares without any explicit prediction elements. Far off the mark, cause the markets where to small/nonexpert. It gets worse with options. Our simulation long ago told us that for sure. That is where the company usage problem is. You should have a very large (international) or very expert company or the pm will only simpleextrapolate (add up) the now or sad missed trends of the past.

Please log in (you may want to register first) to post comments!