I did not make this headline up
- Member for
- 4 years 9 weeks
- View recent blog entries
|5 days 11 hours ago||MGoQuestion||
I was expecting you to use the MGoQuestion from Mattison's presser last week?
|1 week 6 days ago||many laws are immoral||
think early voting laws.
|2 weeks 6 days ago||Baumgardner needs an editor||
"Michigan's incredibly unique and one-of-a-kind football coach"
|6 weeks 4 days ago||best name?||
good post, but not 100% comfortable with,this new "let's laugh at the crazy African american names" tradition.
|7 weeks 3 days ago||fantastic read||
Thanks for sharing!
|7 weeks 5 days ago||double post||
|7 weeks 5 days ago||It's not about what Tom Brady wants to drive||
It's about what Average Joe who watches Tom Brady get a truck might want to buy.
|7 weeks 6 days ago||tongue in cheek||
Well done on the footnote. You know your readers well :)
|8 weeks 2 days ago||well played||
|8 weeks 3 days ago||teenage daughter?||
Get her to watch this: http://www.theverge.com/2015/1/20/7862133/run-the-jewels-killer-mike-elp-love-advice-video
|8 weeks 4 days ago||That is, zero sense if you||
That is, zero sense if you think about it as an investment in human capital. If you think of your education as a "consumption good", and that he truly enjoys the experience that much more, that's a different matter.
|8 weeks 4 days ago||Labor economics says "no"||
There is a lot of serious research in labor economics which suggests that, for most students, the whole undergraduate degree is not even worth that much. Doubling the cost makes zero sense.
|12 weeks 5 days ago||Wow, this looks like shit||
Sorry, I like Hackett and all, but this watch looks like crap.
|14 weeks 14 hours ago||Center not study||
This is not for one study. Faculty salaries, grad student support, organizing/hosting conferences (expensive), and then actual research support (also expensive).
Great scholars at that center. Well deserved.
|14 weeks 1 day ago||Simply beautiful.||
Simply beautiful. You're the man.
|16 weeks 4 days ago||reference||
Can't figure out if that's a reference W's big brown beaver or to Southbound pachyderm
|17 weeks 9 hours ago||saxophone|
|17 weeks 6 days ago||deleted||
|17 weeks 6 days ago||Won't reach the goal||
The website says:This campaign launched on Nov 7, 2014 and will only receive funds if it reaches its goal by Feb 11, 2015."
The goal is $1,000,000...
|18 weeks 3 days ago||"The Evolution of||
"The Evolution of Cooperation" (with William Hamilton) Science, 211, (27 March 1981)
“Evolution of Cooperation without Reciprocity”with Rick L.Riolo and Michael D.
Cohen, Nature, 414 (22 November 2001), pp. 441-443.
|18 weeks 3 days ago||Very rare||
He's probably one of the top 4 or 5 most cited political scientists and he has published in outlets like Science, I think. So that's why.
|18 weeks 3 days ago||Bob is great||
|21 weeks 6 days ago||I sure hope so||
I sure hope so
|25 weeks 3 hours ago||Gambler's fallacy||
Look it up
|26 weeks 12 hours ago||Well played||
Well player, sir
|27 weeks 4 days ago||stats.ncaa.org||
Probably has what you're looking for.
Look for football and game by game results.
|29 weeks 3 days ago||Bayesian updating||
Starting odds of victory at 50% for every game doesn't strike me as particularly reasonable. Could you build in some bayesian updating to your model, perhaps using Vegas odds as your prior?
|32 weeks 2 days ago||yes, that's right. One||
yes, that's right.
One way to think about this is that the initial regression results were estimated using the smallest bins possible (i.e., one bin per observation). By binning, we artificially boost the R^2, and also introduce some bias in the regression estimates. This suggests that the first model should be the one we use to predict, not the last model (though we probably shouldn't put too much stock in these predictions anyway).
For DG, we have:
78.564 - .5677 × 146.1 = -4.37697
|32 weeks 2 days ago||Binning and R^2||
I don't think your binning/averaging approach works. It only compresses the data around the best fit line, which simply reduces the variance to be explained. Given the definition of R^2, it’s not surprising at all to see its value jump near 1.
If you’re worried about outliers and high leverage points, try using some form of robust regression (e.g., M-estimator). Though to be honest, I’m really not sure that would be worth it: to my eye, the plots don’t seem to reveal any huge outlier issues.
Overall, I would say a .31 R^2 is actually pretty impressive given that this is just a bivariate relationship.
|1 year 8 weeks ago||Same reaction||
I don't comment often but I have to say I had the same reaction. This was a careless and insensitive post. This website is an important platform for you, and that comes with responsibility.