Andrew Gelman

I just squared it there because I wanted to create a simple example where x does not have any simple distribution. I don't remember exactly why, but I wanted an...

Fixed; thanks.

I'm not an expert on the Lalonde example; that's Jennifer's domain. But I thought the whole point of Lalonde was that they had an observational study and then also an...

It's a bug, something to do with R, it's getting fixed.

We discuss this in BDA in the simulation chapter, the idea that even direct simulations are noisy! > On Aug 5, 2017, at 10:42 AM, Bob Carpenter wrote: > In...

Hi, thanks for thinking of me! For Waic and cross-validation we will want to evaluate individual terms of the log likelihood but for mle etc., we just need to evaluate...

Setting aside sparse matrices for a moment, I just wanted to say that, yes, I understand the conceptual difference between vectors and arrays. But when working in practice with Stan,...

Things I can do with a vector or matrix but not a 1-d or 2-d real array: - vector and matrix operations (multiplication etc) - various functions that are vectorized....

At the very least, I'd extend to f(x) = f(x_1) below the minimum value and f(x) = f(x_N) above the maximum value (or, if we go with some smoother curve,...

I put in a vote for piecewise linear, at least as one of the options. I say this first because it's quick and transparent, second because I've done this linear...