andrew_gelman_stats andrew_gelman_stats-2012 andrew_gelman_stats-2012-1540 knowledge-graph by maker-knowledge-mining
Source: html
Introduction: David Pennock writes: http://PredictWiseQ.com is our (beta) prediction contest which aims to estimate not just the marginal probabilities of election outcomes this November, but millions of correlations among outcomes as well, like the chance Obama will win both Ohio and Florida, or the chance Romney will win if the September jobs numbers are negative. It’s a working example of a combinatorial prediction market design we published this summer in the conference ACM EC’12. And here’s Pennock’s blog, which supplies more background.
sentIndex sentText sentNum sentScore
1 It’s a working example of a combinatorial prediction market design we published this summer in the conference ACM EC’12. [sent-3, score-0.929]
2 And here’s Pennock’s blog, which supplies more background. [sent-4, score-0.202]
wordName wordTfidf (topN-words)
[('pennock', 0.479), ('win', 0.242), ('acm', 0.24), ('prediction', 0.23), ('outcomes', 0.228), ('contest', 0.209), ('supplies', 0.202), ('ohio', 0.185), ('chance', 0.178), ('aims', 0.174), ('florida', 0.165), ('beta', 0.162), ('romney', 0.157), ('september', 0.154), ('november', 0.15), ('summer', 0.147), ('conference', 0.144), ('millions', 0.136), ('marginal', 0.131), ('market', 0.129), ('correlations', 0.128), ('jobs', 0.127), ('probabilities', 0.12), ('obama', 0.116), ('http', 0.108), ('election', 0.108), ('background', 0.104), ('design', 0.098), ('david', 0.088), ('numbers', 0.083), ('among', 0.082), ('estimate', 0.075), ('working', 0.072), ('published', 0.069), ('blog', 0.053), ('well', 0.05), ('example', 0.04), ('writes', 0.036), ('like', 0.026)]
simIndex simValue blogId blogTitle
same-blog 1 1.0 1540 andrew gelman stats-2012-10-18-“Intrade to the 57th power”
Introduction: David Pennock writes: http://PredictWiseQ.com is our (beta) prediction contest which aims to estimate not just the marginal probabilities of election outcomes this November, but millions of correlations among outcomes as well, like the chance Obama will win both Ohio and Florida, or the chance Romney will win if the September jobs numbers are negative. It’s a working example of a combinatorial prediction market design we published this summer in the conference ACM EC’12. And here’s Pennock’s blog, which supplies more background.
Introduction: I recently wrote about the difficulty people have with probabilities, in this case the probability that Obama wins the election. If the probability is reported as 70%, people think Obama is going to win. Actually, though, it just means that Obama is predicted to get about 50.8% of the two-party vote, with an uncertainty of something like 2 percentage points. So, as I wrote, the election really is too close to call in the sense that the predicted vote margin is less than its uncertainty. But . . . when people see a number such as 70%, they tend to attribute too much certainty to it. Especially when the estimated probability has increased from, say 60%. How to get the point across? Commenter HS had what seems like a good suggestion: Say that Obama will win, but there is 25% chance (or whatever) that this prediction is wrong? Same point, just slightly different framing, but somehow, this seems far less incendiary. I like that. Somehow a stated probability of 75% sounds a
3 0.16461745 1547 andrew gelman stats-2012-10-25-College football, voting, and the law of large numbers
Introduction: In an article provocatively entitled, “Will Ohio State’s football team decide who wins the White House?”, Tyler Cowen and Kevin Grier report : It is statistically possible that the outcome of a handful of college football games in the right battleground states could determine the race for the White House. Economists Andrew Healy, Neil Malhotra, and Cecilia Mo make this argument in a fascinating article in the Proceedings of the National Academy of Science. They examined whether the outcomes of college football games on the eve of elections for presidents, senators, and governors affected the choices voters made. They found that a win by the local team, in the week before an election, raises the vote going to the incumbent by around 1.5 percentage points. When it comes to the 20 highest attendance teams—big athletic programs like the University of Michigan, Oklahoma, and Southern Cal—a victory on the eve of an election pushes the vote for the incumbent up by 3 percentage points. T
4 0.1235133 1569 andrew gelman stats-2012-11-08-30-30-40 Nation
Introduction: Barack Obama’s win has a potentially huge effect on policy. The current budget negotiations will affect the level and direction of government spending and on the mix of taxes paid by different groups of Americans. We can guess that a President Romney would have fought hard against upper-income tax increases. Other areas of long-term impact include the government’s stance on global warming, foreign policy, and the likelihood that Obama will nominate new Supreme Court justices who will uphold the right to abortion announced in Roe v. Wade. When it comes to public opinion, the story is different. The Democrats may well benefit in 2014 and 2016 from the anticipated slow but steady recovery of the economy over the next few years—but, as of November 6, 2012, the parties are essentially tied, with Barack Obama receiving 51% of the two-party vote, compared to Mitt Romney’s 49%, a split comparable to Al Gore’s narrow victory in 2000, Richard Nixon’s in 1968, and John Kennedy’s in 1960.
5 0.1169445 2216 andrew gelman stats-2014-02-18-Florida backlash
Introduction: In a post entitled, “A holiday message from the creative class to Richard Florida — screw you,” Mark Palko argues that Florida’s famous theories about the rise of the creative class have not held up over time: Florida paints a bright picture of these people and their future, with rapidly increasing numbers, influence and wealth. He goes so far as to say “Places that succeed in attracting and retaining creative class people prosper; those that fail don’t.” . . . But, Palko argues, Except for a few special cases, this may be the worst time to make a living in the arts since the emergence of modern newspapers and general interest magazines and other mass media a hundred and twenty years ago . . . Though we now have tools that make creating and disseminating art easier than ever, no one has come up with a viable business model that supports creation in today’s economy. . . . OK, fine, so individual creatives aren’t doing so well? But what about the larger urban economies? P
7 0.11234142 1983 andrew gelman stats-2013-08-15-More on AIC, WAIC, etc
9 0.1042642 607 andrew gelman stats-2011-03-11-Rajiv Sethi on the interpretation of prediction market data
10 0.099898674 2262 andrew gelman stats-2014-03-23-Win probabilities during a sporting event
11 0.097629704 1075 andrew gelman stats-2011-12-20-This guy has a regular column at Reuters
12 0.09527389 1242 andrew gelman stats-2012-04-03-Best lottery story ever
13 0.091295779 1742 andrew gelman stats-2013-02-27-What is “explanation”?
14 0.091117948 279 andrew gelman stats-2010-09-15-Electability and perception of electability
15 0.089862421 1297 andrew gelman stats-2012-05-03-New New York data research organizations
16 0.083232157 1237 andrew gelman stats-2012-03-30-Statisticians: When We Teach, We Don’t Practice What We Preach
17 0.080103591 391 andrew gelman stats-2010-11-03-Some thoughts on election forecasting
18 0.079037964 1574 andrew gelman stats-2012-11-12-How to Lie With Statistics example number 12,498,122
19 0.076132976 1911 andrew gelman stats-2013-06-23-AI Stats conference on Stan etc.
20 0.075985655 84 andrew gelman stats-2010-06-14-Is it 1930?
topicId topicWeight
[(0, 0.075), (1, -0.019), (2, 0.068), (3, 0.043), (4, -0.036), (5, 0.022), (6, -0.022), (7, -0.035), (8, -0.037), (9, -0.017), (10, 0.009), (11, 0.022), (12, 0.02), (13, -0.045), (14, -0.079), (15, 0.014), (16, -0.01), (17, 0.017), (18, 0.006), (19, 0.04), (20, 0.016), (21, 0.081), (22, 0.043), (23, 0.077), (24, 0.005), (25, 0.037), (26, -0.013), (27, 0.039), (28, 0.011), (29, -0.051), (30, -0.02), (31, 0.026), (32, -0.022), (33, 0.002), (34, -0.015), (35, -0.006), (36, 0.038), (37, -0.008), (38, 0.006), (39, -0.039), (40, -0.025), (41, -0.034), (42, 0.04), (43, 0.012), (44, 0.015), (45, 0.024), (46, 0.007), (47, 0.053), (48, -0.081), (49, -0.045)]
simIndex simValue blogId blogTitle
same-blog 1 0.98173183 1540 andrew gelman stats-2012-10-18-“Intrade to the 57th power”
Introduction: David Pennock writes: http://PredictWiseQ.com is our (beta) prediction contest which aims to estimate not just the marginal probabilities of election outcomes this November, but millions of correlations among outcomes as well, like the chance Obama will win both Ohio and Florida, or the chance Romney will win if the September jobs numbers are negative. It’s a working example of a combinatorial prediction market design we published this summer in the conference ACM EC’12. And here’s Pennock’s blog, which supplies more background.
2 0.77510899 1556 andrew gelman stats-2012-11-01-Recently in the sister blogs: special pre-election edition!
Introduction: 1. I can simultaneously (a) accept that Obama has a 72 percent chance of winning and (b) say the election is too close to call 2. Michael’s a Republican, Susan’s a Democrat 3. The narcissism of the narcissism of small differences 4. Obamanomics: A Counter-counterhistory 5. Not a gaffe 6. Categories influence predictions about individual consistency
Introduction: The other day we had a fun little discussion in the comments section of the sister blog about the appropriateness of stating forecast probabilities to the nearest tenth of a percentage point. It started when Josh Tucker posted this graph from Nate Silver : My first reaction was: this looks pretty but it’s hyper-precise. I’m a big fan of Nate’s work, but all those little wiggles on the graph can’t really mean anything. And what could it possibly mean to compute this probability to that level of precision? In the comments, people came at me from two directions. From one side, Jeffrey Friedman expressed a hard core attitude that it’s meaningless to give a probability forecast of a unique event: What could it possibly mean, period, given that this election will never be repeated? . . . I know there’s a vast literature on this, but I’m still curious, as a non-statistician, what it could mean for there to be a meaningful 65% probability (as opposed to a non-quantifiab
4 0.7315374 1574 andrew gelman stats-2012-11-12-How to Lie With Statistics example number 12,498,122
Introduction: This post is by Phil Price. Bill Kristol notes that “Four presidents in the last century have won more than 51 percent of the vote twice: Roosevelt, Eisenhower, Reagan and Obama”. I’m not sure why Kristol, a conservative, is promoting the idea that Obama has a mandate, but that’s up to him. I’m more interested in the remarkable bit of cherry-picking that led to this “only four presidents” statistic. There was one way in which Obama’s victory was large: he won the electoral college 332-206. That’s a thrashing. But if you want to claim that Obama has a “popular mandate” — which people seem to interpret as an overwhelming preference of The People such that the opposition is morally obligated to give way — you can’t make that argument based on the electoral college, you have to look at the popular vote. That presents you with a challenge for the 2012 election, since Obama’s 2.7-point margin in the popular vote was the 12th-smallest out of the 57 elections we’ve had. There’s a nice sor
Introduction: I recently wrote about the difficulty people have with probabilities, in this case the probability that Obama wins the election. If the probability is reported as 70%, people think Obama is going to win. Actually, though, it just means that Obama is predicted to get about 50.8% of the two-party vote, with an uncertainty of something like 2 percentage points. So, as I wrote, the election really is too close to call in the sense that the predicted vote margin is less than its uncertainty. But . . . when people see a number such as 70%, they tend to attribute too much certainty to it. Especially when the estimated probability has increased from, say 60%. How to get the point across? Commenter HS had what seems like a good suggestion: Say that Obama will win, but there is 25% chance (or whatever) that this prediction is wrong? Same point, just slightly different framing, but somehow, this seems far less incendiary. I like that. Somehow a stated probability of 75% sounds a
6 0.72362024 1075 andrew gelman stats-2011-12-20-This guy has a regular column at Reuters
7 0.69636214 1569 andrew gelman stats-2012-11-08-30-30-40 Nation
8 0.64695108 934 andrew gelman stats-2011-09-30-Nooooooooooooooooooo!
10 0.60604 2262 andrew gelman stats-2014-03-23-Win probabilities during a sporting event
11 0.5671261 292 andrew gelman stats-2010-09-23-Doug Hibbs on the fundamentals in 2010
12 0.56493247 84 andrew gelman stats-2010-06-14-Is it 1930?
13 0.55359572 1567 andrew gelman stats-2012-11-07-Election reports
14 0.54703832 473 andrew gelman stats-2010-12-17-Why a bonobo won’t play poker with you
15 0.54457092 966 andrew gelman stats-2011-10-20-A qualified but incomplete thanks to Gregg Easterbrook’s editor at Reuters
16 0.5423364 551 andrew gelman stats-2011-02-02-Obama and Reagan, sitting in a tree, etc.
17 0.53365439 138 andrew gelman stats-2010-07-10-Creating a good wager based on probability estimates
19 0.51508278 628 andrew gelman stats-2011-03-25-100-year floods
20 0.50974512 337 andrew gelman stats-2010-10-12-Election symposium at Columbia Journalism School
topicId topicWeight
[(0, 0.229), (21, 0.037), (22, 0.023), (24, 0.127), (34, 0.024), (86, 0.093), (93, 0.043), (95, 0.045), (99, 0.237)]
simIndex simValue blogId blogTitle
same-blog 1 0.93647802 1540 andrew gelman stats-2012-10-18-“Intrade to the 57th power”
Introduction: David Pennock writes: http://PredictWiseQ.com is our (beta) prediction contest which aims to estimate not just the marginal probabilities of election outcomes this November, but millions of correlations among outcomes as well, like the chance Obama will win both Ohio and Florida, or the chance Romney will win if the September jobs numbers are negative. It’s a working example of a combinatorial prediction market design we published this summer in the conference ACM EC’12. And here’s Pennock’s blog, which supplies more background.
2 0.91666138 565 andrew gelman stats-2011-02-09-Dennis the dentist, debunked?
Introduction: Devah Pager points me to this article by Uri Simonsohn, which begins: Three articles published [by Brett Pelham et al.] have shown that a disproportionate share of people choose spouses, places to live, and occupations with names similar to their own. These findings, interpreted as evidence of implicit egotism, are included in most modern social psychology textbooks and many university courses. The current article successfully replicates the original findings but shows that they are most likely caused by a combination of cohort, geographic, and ethnic confounds as well as reverse causality. From Simonsohn’s article, here’s a handy summary of the claims and the evidence (click on it to enlarge): The Pelham et al. articles have come up several times on the blog, starting with this discussion and this estimate and then more recently here . I’m curious what Pelham and his collaborators think of Simonsohn’s claims.
3 0.89323068 2166 andrew gelman stats-2014-01-10-3 years out of date on the whole Dennis the dentist thing!
Introduction: Paging Uri Simonsohn . . . January 2014: Alice Robb writes , completely uncritically: “If Your Name is Dennis, You’re More Likely to Become a Dentist The strange science of how names shape careers.” But look what you can learn from a quick google: Hmmmm, maybe worth following up on that second link . . . More details here , from 2011: Devah Pager points me to this article by Uri Simonsohn, which begins: Three articles published [by Brett Pelham et al.] have shown that a disproportionate share of people choose spouses, places to live, and occupations with names similar to their own. These findings, interpreted as evidence of implicit egotism, are included in most modern social psychology textbooks and many university courses. The current article successfully replicates the original findings but shows that they are most likely caused by a combination of cohort, geographic, and ethnic confounds as well as reverse causality. From Simonsohn’s article, here’s a han
4 0.89241862 1068 andrew gelman stats-2011-12-18-Faculty who don’t like teaching and hate working with students
Introduction: Peter Woit links to Steve Hsu linking to a 1977 interview by Katherine Sopka of physics professor Sidney Coleman. I don’t know anything about Coleman’s research but the interview caught my eye because one of my roommates in grad school was one of Coleman’s advisees. Anyway, here’s the key bit from the interview: Sopka: But you do enjoy working with students or do you? Coleman: No. I hate it. You do it as part of the job. Well, that’s of course false…or maybe more true than false when I say I hate it. Occasionally there’s a student who is a joy to work with. But I certainly would be just as happy if I had no graduate students… Sopka: I guess your remark means then that you would like to avoid teaching undergraduate courses or even required graduate courses… Coleman: Or even special topics courses. Teaching is unpleasant work. No question about it. It has its rewards. One feels happy about having a job well done. Washing the dishes, waxing the floors (things I also
5 0.87866575 1736 andrew gelman stats-2013-02-24-Rcpp class in Sat 9 Mar in NYC
Introduction: Join Dirk Eddelbuettel for six hours of detailed and hands-on instructions and discussions around Rcpp, RInside, RcppArmadillo, RcppGSL and other packages . . . Rcpp has become the most widely-used language extension for R. Currently deployed by 103 CRAN packages and a further 10 BioConductor packages, it permits users and developers to pass “whole R objects” with ease between R and C++ . . . Morning session: “A Hands-on Introduction to R and C++” . . . Afternoon session: “Advanced R and C++ Topics” . . .
6 0.87041432 1828 andrew gelman stats-2013-04-27-Time-Sharing Experiments for the Social Sciences
7 0.83415473 381 andrew gelman stats-2010-10-30-Sorry, Senator DeMint: Most Americans Don’t Want to Ban Gays from the Classroom
8 0.82703435 1075 andrew gelman stats-2011-12-20-This guy has a regular column at Reuters
9 0.82125044 4 andrew gelman stats-2010-04-26-Prolefeed
10 0.8179794 1781 andrew gelman stats-2013-03-29-Another Feller theory
11 0.81773829 40 andrew gelman stats-2010-05-18-What visualization is best?
12 0.81747425 365 andrew gelman stats-2010-10-24-Erving Goffman archives
13 0.81074619 557 andrew gelman stats-2011-02-05-Call for book proposals
14 0.80819881 1177 andrew gelman stats-2012-02-20-Joshua Clover update
15 0.80588025 862 andrew gelman stats-2011-08-20-An illustrated calculus textbook
16 0.78958076 1765 andrew gelman stats-2013-03-16-Recently in the sister blog
17 0.78387022 2173 andrew gelman stats-2014-01-15-Postdoc involving pathbreaking work in MRP, Stan, and the 2014 election!
18 0.78140676 1782 andrew gelman stats-2013-03-30-“Statistical Modeling: A Fresh Approach”
19 0.77199537 2190 andrew gelman stats-2014-01-29-Stupid R Tricks: Random Scope
20 0.76327682 20 andrew gelman stats-2010-05-07-Bayesian hierarchical model for the prediction of soccer results