The authors are currently working on the next iteration of the model which will hopefully come with a few improvements.
AFAIK, they are keen on getting their hands on data that proved difficult for the model -- I suggest you get in touch with them if you think you could help!
My big disappointment is that the model does not provide sources and recommended reading. Which is something we can now do and would increase the usefulness of the model significantly.
From the title I was hoping this would talk a little more about the model used. At 45 minutes runtime, I'm guessing it's something more interesting than a basic linear model.
Are these gathered in one place anywhere? I and probably many others, including the authors of this paper, would be interested in these as a test set for models like this.
Weird, in '08 he had a page giving the model in detail, including some of the fit values (e.g. the imputed pollster "house effects"), but I can't find it this time around. Change since the NYT purchase?
This is really cool -- thanks for posting it! I'll have to skim through it at some point since a lot of my work is in classifications models and mirrors the results you've seen
reply