tag:blogger.com,1999:blog-1369432396898204613.post2982707769856953437..comments2023-01-15T10:39:00.543+01:00Comments on Lighthouse in the Sky: PredictionUnknownnoreply@blogger.comBlogger5125tag:blogger.com,1999:blog-1369432396898204613.post-55799267418371788512012-12-30T15:21:12.829+01:002012-12-30T15:21:12.829+01:00also: http://mathbabe.org/2012/12/20/nate-silver-c...also: http://mathbabe.org/2012/12/20/nate-silver-confuses-cause-and-effect-ends-up-defending-corruption/mvchttps://www.blogger.com/profile/14404841243741347001noreply@blogger.comtag:blogger.com,1999:blog-1369432396898204613.post-39436744713628724382012-11-26T16:53:25.699+01:002012-11-26T16:53:25.699+01:00I just finished the book. Fascinating indeed!
Wha...I just finished the book. Fascinating indeed!<br /><br />What I found most interesting was the chapter on global warming. While it's true that both 'sides' of the issue commit statistic felony, it's clear who's backing the better bet. And Silver manages to show it using only a limited amount of handwaving.Popupnoreply@blogger.comtag:blogger.com,1999:blog-1369432396898204613.post-46766240181065189202012-11-13T15:00:47.881+01:002012-11-13T15:00:47.881+01:00anecdote: when my brother worked as a snow plow dr...anecdote: when my brother worked as a snow plow driver, he was (understandably) an obsessive weather forecast reader. over time, he came to prefer environment canada, saying that commercial forecasters would generally overestimate the probability of good weather three or four days into the future, then gradually downgrade their forecast as a storm approached. apparently (they believe) consumers want more bright sunshine icons in their forecasts, even at the expense of accuracy.<br /><br />and speaking of following things obsessively, this cartoon summarizes why i was reading nate silver during the last two US presidential elections: http://cdn.pjmedia.com/vodkapundit/files/2012/11/121029_daily-cartoon-silver_p600.jpg<br /><br />too bad no-one can pull that trick off here (much harder election system to simulate, and far less data available). silver himself did a rather poor job with the last general election in the UK, for example.mvchttps://www.blogger.com/profile/14404841243741347001noreply@blogger.comtag:blogger.com,1999:blog-1369432396898204613.post-58401137529392992932012-11-12T10:53:40.118+01:002012-11-12T10:53:40.118+01:00Hmm, no I haven't read either. Perhaps it'...Hmm, no I haven't read either. Perhaps it's time to visit the library. <br /><br />From their descriptions, though, they seem a little more focused on people's reactions, whereas Silver's book is aimed squarely at making accurate predictions. People's reactions and motivations are simply obstacles to this goal, though they're unavoidable obstacles, and he talks about ways to try to make sure the incentives for forecasters lead to accurate predictions. The classic, of course, is markets and betting - a more-accurate forecaster can always make money in the long run by betting against a less-accurate forecaster. In Physics Experiment Land, that is. In reality, we've seen things go badly wrong with markets, and even markets like InTrade intended specifically for forecasting.Anonymoushttps://www.blogger.com/profile/00764119699293212898noreply@blogger.comtag:blogger.com,1999:blog-1369432396898204613.post-64215874213448011722012-11-12T10:36:19.616+01:002012-11-12T10:36:19.616+01:00Interesting!
This has been on my to-read list for...Interesting!<br /><br />This has been on my to-read list for some time now. Maybe it's time I got on to it.<br /><br />It sounds like a companion to Talebs <a href="https://en.wikipedia.org/wiki/The_Black_Swan_%28Taleb_book%29" rel="nofollow">The Black Swan</a> or Dan Gardners <a href="http://www.amazon.com/Risk-The-Science-Politics-Fear/dp/1905264151?tag=duckduckgo-d-20" rel="nofollow">Risk - the science and politics of Fear</a>.<br /><br />Have you read either book?Popupnoreply@blogger.com