[nullc]

Meh. If you fit a model and don't explicitly constrain against "un-physical" results like negative correlations, you'll end up with them.

Constraining against them won't improve your models fit (usually by definition), and it doesn't always improve robustness (at least for situations near average)-- because they're acting to debias the model in ways that you otherwise don't have enough degrees of freedom to address.

A negative correlation here is also potentially historically supported, in the sense that sometimes DEM/GOP candidates are philosophically reversed in some way relevant to the state. As in, "The only way a GOP would get elected in X is if they had the DEM position on subject Y which would make them lose state Z, who cares as much about that subject as X but in the opposite direction."

Now-- it doesn't seem likely case in this election (e.g. Trump is not (currently) a massively pro-choice republican), so it probably shouldn't apply here-- but it's isn't hard for me to imagine how a negative correlation might show up out of the historical data.

[RivieraKid]

> If you fit a model and don't explicitly constrain against "un-physical" results like negative correlations, you'll end up with them.

The Economist model does exactly that, and all of their correlations are positive.

I recommend reading their methodology, they know what they're doing (I wouldn't say the same about 538). Andrew Gelman has developed some of the Bayesian methods and software that people like Nate Silver use, he's the main author of what's considered a reference book on Bayesian statistics.

[noirbot]

I think the question is if it matters to the predictive accuracy of the model. Just because it puts out results you can't envision actually happening on the margins doesn't mean they can't happen, or that they can't be valuable in presenting a holistic result.

It's clear that the models are tuned differently, but from Silver's replies in the PS's, it seems that he's ok with these artifacts being part of the model.

[RivieraKid]

Yes, it increases the state-level and national uncertainty intervals (Andrew Gelman has talked about this several times on his blog), which improves Trump's odds.

[noirbot]

Sure, but that's not necessarily wrong. Any decision in the model will change Trump's odds in one way or another. The question is if it makes it closer to the (unknowable) real odds.

Just because intuition says it should be longer odds for Trump doesn't mean that's right.

[sangnoir]

My statistics knowledge has withered away, but isn't this quibbling over overall approach? 538 seems do be doing a top-down approach while the Economist is more bottom-up. What is strange is that a person affiliated to the Economist is then asking why the 538 model's emergent properties aren't exhibiting more bottom-up characteristics .