No comments yet

No comments yet

## Search The Blog

## Search

## Recent Comments

- Daniel Lakeland on Injecting Data into the SF Rent discussion: “My own personal belief is that it's all down to the tech bubble taking off as a result of all…” May 20, 13:10
- Morgan on Injecting Data into the SF Rent discussion: “The break in distribution shape between 2011and 2012 seems impossibly large. Any idea what happened there?” May 20, 12:45
- Daniel Lakeland on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “Sure, so long as you define composition in terms of division by the derivative you are able to enforce an…” Apr 21, 12:28
- Corey Yanofsky on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “The indeterminacy of ds1/ds2 goes away because we require that the resulting density be normalized. Compare the scaling property of…” Apr 20, 08:40
- Daniel Lakeland on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “Aren't these basically the same issue? Suppose we use normal(x,M,ds1) with ds1 infinitesimal as a delta function for x in…” Apr 19, 13:16
- Corey Yanofsky on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “Here the generative model approach makes things clear. You have the top-level R and N slightly perturbed versions of it;…” Apr 18, 06:26
- Daniel Lakeland on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “I keep wanting to work on this in some kind of detail but it will have to wait a few…” Apr 18, 05:24
- Corey Yanofsky on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “I think it does -- you end up having to marginalize out the experiment-level R parameters and that leaves you…” Apr 17, 21:53
- Daniel Lakeland on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “Right that makes sense. However I don't think it poses a problem for the case where there is one R…” Apr 17, 11:20
- Corey on Inference on a ratio from two separate experiments (a case study in Stan awesomeness): “This issue isn't just that the infinitesimal elements in normal_pdf(x/y-R,ds1) and normal_pdf(p/q-R,ds2) don't have a well-defined ratio. It's that for…” Apr 17, 06:18

## Categories

## The Archives

## Tags

applied mathematics baby bayesian statistics beams Biology biomechanics computer science declarative models differential geometry Energy Efficiency example problems falling ball fluid mechanics golf granular materials health insurance heat conduction HVAC lagrangian mechanics liquefaction lit review mcmc mechanics Medicine modeling nonlinear dynamics nonstandard analysis optimization PDE philosophy of science physics polynomial chaos poverty trap radial basis functions randomness renormalization group statistics Statistics In Science taxes Teaching turkeys ubi universal basic income utility theory zombies

Copyright 2017