HN Explorer

SAS is looking for a Bayesian Specialist (sas.com)

4 points by pjonesdotca ↗ HN
Description

Join the world's leading statistical software company and make a difference in the way that statistics is practiced! SAS is expanding its statistical procedures by incorporating modern Bayesian methods, and we are looking for a Bayesian specialist to accelerate this exciting development effort.

In this position, you will implement Bayesian inference methods in new and existing SAS procedures. You will select optimal sampling methods for specific Bayesian models and program the methods in C. You will also test and document the software and present it to professional statistical audiences.

Qualifications

Essential

    *
      Ph.D. in statistics, biostatistics, computer science, or a related field.
    *
      Experience with Bayesian statistics.
    *
      Experience with model fitting samplers, such as versions of the Metropolis algorithm and Monte Carlo methods.
    *
      Experience writing large programs in C, C++, FORTRAN, or a similar scientific programming language.
(note) I write MonteCarlo apps in Ruby. Does that make Ruby a "scientific language"?

(note (note)) What do they mean "scientific language"?

2 comments

[ 146 ms ] story [ 85.6 ms ] thread
pjonesdotca
I have and still write MonteCarlo/Bootstrapping apps in Ruby. Does that make Ruby a "scientific language"?

What do they mean "scientific language" anyways?

joshu
SAS has a boatload of statistical primitives. I dunno what makes it "scientific" but it's way useful for doing that kind of thing.

(I did SAS for years, building trading models. Which isn't science.)

I hated the language, but I miss the functionality terribly.

You CAN do this sort of work. But it's nice for k-means or whatever to be a one-liner.