南京老萝卜 发表于 2016-5-8 10:02
希拉里是72.2%，川普是23.5%，优势超过3：1，这希拉里不是横扫川普如卷席么？
不知道这个调查是否靠谱。 ...
这个不是调查。而是根据各个betting market上给出的赔率，综合换算出的获胜概率。也就是说压希拉里3块钱的话，如果她输了就没了，如果赢了能赚1块钱多一点。压Trump 1块钱的话，如果他赢了，能够赚2块钱多。因此我说如果爱坛坛友有认为Trump赢的可能性非常大的话，这是赚钱的好机会。在betting market上下注的是自己的真金白银，会比较理性，受感情因素干扰不大，赔率是根据诸多的下注结果进行调整的，可以说综合了各方面的信息。经济学家做过一些研究，betting market在各种预测方法中是很准的。
好吧，我开了个不正经楼，欢迎围观。。。
http://www.aswetalk.net/bbs/forum.php?mod=viewthread&tid=42090
Dracula 发表于 2016-5-8 13:17
这个不是调查。而是根据各个betting market上给出的赔率，综合换算出的获胜概率。也就是说压希拉里3块钱 ...
这种下注有emotional hedging的因素。
比如说，我看好DT赢，但是我会赌DT输。如果DT赢了，皆大欢喜。如果DT输了，我不至于太难过，因为还有赢了点钱。这样下注部分回避了情感损失的风险。
老马丁 发表于 2016-5-9 04:44
这种下注有emotional hedging的因素。
比如说，我看好DT赢，但是我会赌DT输。如果DT赢了，皆大欢喜。如 ...
现实中有几个人是这么下注的呢？比如像体育比赛押注，有几个球迷会因为这个理由买对手球队赢呢？我对棒球比较熟，从来没有听说过有大批的Red Sox 球迷买Yankees赢。而且如果你比较相信efficient market的话，即使有一部分市场参与者不理性，也并不意味价格会很大偏离其基本价值，会有别的理性参与者进场来利用这一点获利，从而将价格推回到其fundamental value。当然prediction market也不是绝对准确，就好像股票市场的估值也会有偏差，有泡沫。但是和其它的各种预测方法相比，betting market的准确度还是很高的。
南京老萝卜 发表于 2016-5-7 18:02
希拉里是72.2%，川普是23.5%，优势超过3：1，这希拉里不是横扫川普如卷席么？
不知道这个调查是否靠谱。 ...
这个不是预测得票率，而是赌胜率，大赢小赢都是赢
Dracula 发表于 2016-5-7 13:09
最新Betting Market，希拉里获胜的概率是72.2%，Sanders是2.4%，Biden是1.2%。加在一起是75.8%。Trump是23. ...
为什么Biden还在上面？而且居然还有market，难道真有人相信选民会一起write in选Biden?
holycow 发表于 2016-5-9 07:41
为什么Biden还在上面？而且居然还有market，难道真有人相信选民会一起write in选Biden? ...
我估计这是在赌如果希拉里在民主党代表大会召开前被FBI起诉，或者得重病，或者甚至被暗杀的话，Biden是有可能成为民主党候选人的。这个概率是百分之一我觉得也差不多。
老马丁 发表于 2016-5-9 04:44
这种下注有emotional hedging的因素。
比如说，我看好DT赢，但是我会赌DT输。如果DT赢了，皆大欢喜。如 ...
上届世界杯的时候我就是这么押注的，我不喜欢德国队，所以买了好几场德国队赢，结果全是平局，一毛没赚到，还没耽误德国队夺冠，我了个擦{:198:}
本帖最后由 Dracula 于 2016-5-10 02:11 编辑
Who will follow Trump off the cliff?
By George F. Will
https://www.washingtonpost.com/opinions/who-will-follow-trump-off-the-cliff/2016/05/06/752678a6-12dc-11e6-93ae-50921721165d_story.html
Donald Trump: “We’ve got to get rid of the $19 trillion in debt.”
The Post: “How long would that take?” . . .
Trump: “I would say over a period of eight years.”
— March 31
Fortune: “You’ve said you plan to pay off the country’s debt in 10 years. How’s that possible?”
Trump: “No, I didn’t say 10 years.”
— April 19
Speaking on “Fox & Friends,” of course, Trump revealed something he learned from the National Enquirer, of course. Although the Kennedy assassination is one of history’s most minutely studied events, all previous scrutiny missed something the supermarket tabloid discovered for people like Trump — a connection between Ted Cruz’s father and the murder of the 35th president. Trump said:
“You know, father was with Lee Harvey Oswald prior to Oswald’s being, you know, shot. I mean the whole thing is ridiculous. What is this, right prior to his being shot, and nobody even brings it up. I mean they don’t even talk about that. That was reported, and nobody talks about it. But I think it’s horrible. I think it’s absolutely horrible that a man can go and do that, what he’s saying there.”
Fox host: “Right. There was a picture out there that reportedly shows Rafael Cruz standing with Lee Harvey Oswald —”
Trump: “I mean what was he doing with Lee Harvey Oswald shortly before the death, before the shooting? It’s horrible.”
Fox host: “Crazy.”
Parsing Trump sentences is a challenge but is rewarding because it frequently reveals that he actually has said nothing at all. When silence descends, there lingers in the air only gauzy innuendo. What did Trump really say about “the whole thing” of Oswald?
Looking on the bright side — speaking of challenges — Trump’s nomination might have two salutary effects: It might counteract “The Cult of the Presidency,” as explored in Gene Healy’s 2008 book with that title. And it might reacquaint Republicans with the reality principle — the need to assess and adapt to facts.
Healy analyzes the delusion of “redemption through presidential politics.” The infantilization of America is apparent in the presidency becoming a semi-sacerdotal office, one that conflates spiritual yearnings and magical thinking about wonders performed by executive power. Trump, with his coarse character and fanciful promises, is an antidote to such superstitions.
Now, regarding realities: In 2012, 93 percent of self-described Republicans who voted did so for Mitt Romney. Trump probably cannot receive 80 percent of what probably will be, because of discouragement and revulsion, a smaller Republican turnout. Romney lost 73 percent of the Hispanic vote; Trump is viewed unfavorably by 82 percent of Hispanics and very unfavorably by 62 percent. Trump probably will receive significantly less than Romney’s ruinous 27 percent of this vote. And because of demographic trends and Trump’s motivating policies and insults, Hispanic turnout probably will be significantly larger than in 2012, as the white percentage of the electorate continues to shrink. Romney won just 37 percent of young voters (18-29); Trump is unlikely even to match this.
Although Romney won 53 percent of married women, he received just 44 percent of the total female vote. Today, Trump trails Hillary Clinton among women by 19 points (35 percent to 54 percent), and most women probably do not yet know that he testifies to the excellence of his penis. (“My fingers are long and beautiful, as, has been well-documented, are various other parts of my body.”) Or that his idea of masculinity is to boast about conquests of women “often seemingly very happily married” and that “I have been able to date (screw).” Or that he says “it doesn’t really matter what write as long as you’ve got a young and beautiful piece of ass.”
In receiving, so far, the support of 4.7 percent of America’s eligible voters, Trump has won a mere plurality of votes in a party approved by only 33 percent of the electorate. This electorate had about 5 percent more Democrats than Republicans even before Trump further tarnished the GOP brand. So, Republicans need to carry independents by more than Romney’s five points. Even in states that have voted Republican since 2000, Trump is viewed unfavorably by 62 percent and strongly unfavorably by 52 percent.
His metabolic urge to be scabrous guarantees that Republican candidates everywhere will be badgered by questions about what they think about what he says. What they say will determine how many of them lose with him, and how many deserve to.
本帖最后由 Dracula 于 2016-5-10 02:11 编辑
老马丁 发表于 2016-5-9 04:44
这种下注有emotional hedging的因素。
比如说，我看好DT赢，但是我会赌DT输。如果DT赢了，皆大欢喜。如 ...
讨论问题就是的了，干嘛要侮辱人呢？
关于emotional hedging 的普遍性，我搜到一篇文章，全文我看不到，只找到abstract。
Reluctance to Hedge Desired Outcomes
Carey Morewedge, Carnegie Mellon University, USA
I examined if people engage in emotional hedging when desired outcomes are uncertain. A substantial percentage of US voters and NFL fans refused to hedge the 2012 Presidential election or the victory of their team, respectively, even when it was irrational not to, apparently because people conflate beliefs and desires.
http://www.acrwebsite.org/volumes/v10e/eacr_v10_14131.pdf
坛友里如果有对统计学尤其是预测方面感兴趣的话，这篇文章是挺值得看的。
How I Acted Like A Pundit And Screwed Up On Donald Trump
Dracula 发表于 2016-5-18 08:49
坛友里如果有对统计学尤其是预测方面感兴趣的话，这篇文章是挺值得看的。
How I Acted Like A Pundit And...
统计学什么时候能预测outlier的？他自己承认一开始既没有模型也没有数据，就是在那里主观猜测。照我说有模型有数据也没区别，他讲的那个阶段，有模型几乎肯定也是错的。
Data和Science不应该出现在同一个句子里，和Art倒可以一起用。
holycow 发表于 2016-5-19 02:45
统计学什么时候能预测outlier的？他自己承认一开始既没有模型也没有数据，就是在那里主观猜测。照我说有 ...
经济学里的预测，好多时候的数据还不如他说的那个阶段呢。他的文章里的反思对我还是挺有启发性的。
删除失败 发表于 2016-5-9 14:25
上届世界杯的时候我就是这么押注的，我不喜欢德国队，所以买了好几场德国队赢，结果全是平局，一毛没赚到 ...
你如果每场都押德国赢还是赢的场次多吧
刚看到的很有意思的一段话
Politics unfortunately abounds in shams that must be treated reverentially for every politician who would succeed. If you are the sort of man whose stomach revolts against treating shams reverentially, you will be well advised to stay out of politics altogether and set up as a prophet; your prophecies may perhaps sow good seed for some future harvest. But as a politician you would be impotent. For at any given time the bulk of your countrymen believe firmly and devoutly, not only in various things that are worthy of belief, but also in illusions of one kind and another; and they will never submit to have their affairs managed for them by one who appears not to share in their credulity. … A wise politician will never grudge a genuflexion or a rapture if it is expected of him by the prevalent opinion.
这段话的作者是英国历史学家F. S. Oliver，出处是他的著作 The Endless Adventure: Personalities and Practicalities in Eighteenth Century England，出版于80多年之前。今天看来还是很有洞察力。
方恨少 发表于 2016-5-19 04:12
你如果每场都押德国赢还是赢的场次多吧
德国队赢得比赛都被我巧妙得错过了{:194:}
删除失败 发表于 2016-5-19 11:39
德国队赢得比赛都被我巧妙得错过了
大师，求下期彩票中奖号{:214:}
Dracula 发表于 2016-5-8 05:09
最新Betting Market，希拉里获胜的概率是72.2%，Sanders是2.4%，Biden是1.2%。加在一起是75.8%。Trump是23. ...
最新的betting market，希拉里获胜的概率是66%，Sanders是3.9%，Biden是2.2%，加起来大约是72%。Trump是26.8%，加上其它几个共和党的获胜概率不到28%。同1个月以前相比Trump获胜的概率稍有上升，主要原因是他顺利整合了共和党内的支持，超过了预期。希拉里获胜概率下降部分原因也包括国务院关于她私人电子邮件服务器的报告对她有很严重的批评。但是目前看来希拉里的优势还是很大，她对Trump获胜的odds还是大致在3比1 。
Two new studies find racial anxiety is the biggest driver of support for Trump
Dracula 发表于 2016-6-2 16:35
最新的betting market，希拉里获胜的概率是66%，Sanders是3.9%，Biden是2.2%，加起来大约是72%。Trump是2 ...
我比较常看的realclearpolitics上希拉里领先3.8，这好像是一家比较亲共和党的。
不过现在还有点早吧，也许等到劳动节那时候的数据比较有意义，你觉得呢？