Ron Paul takes the lead in Iowa.

[kentonio]

I don't know **** about markets, but I do follow politics and have for a very long time. When you come into a politics thread saying things like..

romney iowa 85-89, santorum iowa 11-18

..I'm going to assume that that is intrades current odds based on how people are betting. If that's wrong then please enlighten me as to how it actually works.

[KrazyHorse]

Yes, you have now quoted the odds as implied by the market at one point this evening. Please explain what metric you're using to assess the accuracy of this prediction.

Do you have any background in this type of analysis at all? Because I'll let you know, as fair warning, that I do (both as a large part of my PhD thesis as well as professionally).

[kentonio]

Yes, you have now quoted te odds as implied by the market at one point this evening. Please explain what metric you're using to assess the accuracy of this prediction.

The final result of course. What use is a prediction that leads to an incorrect result? I realize this may be horribly simplistic of me, but if it doesn't point to how things are actually going to turn out, whats the point?

[KrazyHorse]

The results are not in yet, as far as I know. In fact, romney iowa 92.5-94.8 santorum iowa 6.0-10.0 is the current quote.

Secondly, you utter imbecile, even in the case that romney were to lose iowa, THE MARKET WAS IMPLYING A 1 IN 7 OR SO CHANCE THAT THIS WOULD HAPPEN. THERE IS ABSOLUTELY NO STATISTICAL TEST YOU COULD DO WITH THOSE TWO PIECES OF INFORMATION WHICH WOULD DEMONSTRATE SIGNIFICANT DEVIATION FROM THE EMH (even if we discount the fact that you cherry-picked a single point in time, which increases the trial factor).

In order to garner information on the accuracy of intrade predictions, what would be required is a larger number of observations of independent events. Now run along and play with the other children

[KrazyHorse]

Looks like romney won. I presume that now based on this unbelievably strong piece of statistical evidence you will admit intrade is incredibly accurate.

[kentonio]

The results are not in yet, as far as I know. In fact, romney iowa 92.5-94.8 santorum iowa 6.0-10.0 is the current quote.

Secondly, you utter imbecile, even in the case that romney were to lose iowa, THE MARKET WAS IMPLYING A 1 IN 7 OR SO CHANCE THAT THIS WOULD HAPPEN. THERE IS ABSOLUTELY NO STATISTICAL TEST YOU COULD DO WITH THOSE TWO PIECES OF INFORMATION WHICH WOULD DEMONSTRATE SIGNIFICANT DEVIATION FROM THE EMH (even if we discount the fact that you cherry-picked a single point in time, which increases the trial factor).

In order to garner information on the accuracy of intrade predictions, what would be required is a larger number of observations of independent events. Now run along and play with the other children

What did that make the odds on Santorum winning? Which it looks like he just did?

[kentonio]

To be accurate shouldn't intrade have been showing fairly equal numbers for both candidates all day?

[KrazyHorse]

Are you actually retarded?

[KrazyHorse]

To be accurate shouldn't intrade have been showing fairly equal numbers for both candidates all day?

Sure; there were 3 possibilities: Paul, Romney, Santorum, therefore each had a 33% chance. That is elementary probability.

[kentonio]

Are you actually retarded?

It's just bull**** isn't it. I think I now have a better insight into why you and your little friends ****ed the world economy a few years back.

You can run along now.

[KrazyHorse]

Yup, becuse you're too dumb to understand even the basics of empirical tests of predictions it means that it's all bull****.

[KrazyHorse]

Do you have some kind of humanities degree by any chance?

[KrazyHorse]

Perhaps filosofy or english literature?

[KrazyHorse]

Savaging you like this feels sort of good, but I get the feeling that I'm hardly the first. Makes it less special somehow.

[KrazyHorse]

Romney now up to 85-86.4 for the nom, Obama down slightly to 50.1-51.3

It looks like there was a substantial "winner" effect on Romney's nomination chances, and a smaller one on the general election (similar to the observed sheepskin effect). Even once it was apparent that the overall numbers would be very close, there was significant comovement between the markets.