6 ms·
I had a stats professor theorize that at some point all stats calculations would be done with simulations instead of formulas, since formulas are simplified mod
by dwater 4y ago
I had a stats professor theorize that at some point all stats calculations would be done with simulations instead of formulas, since formulas are simplified models of reality whereas simulations can capture much more of the complexity if properly constructed. This was 10 or 15 years ago and the feeling was computers were either fast enough already or would be soon enough for most problems. This came up as a result of a similar observation about the transition from tables to graphing calculators for probability distributions and things like that.
- stdbrouw 4y agoPlenty of computing power 15 years ago for bootstrapping (simulation-based estimation of uncertainty for an analytically estimated model), today plenty for the entire model itself (Bayesian MCMC, frequentist optimization-based methods)
- CrazyStat 4y agoIf you read the original 1979 bootstrap paper, one of the delightful things about it is that it discusses the computational cost (literally, in dollars) to rent time for the procedure on a shared mainframe.
- blt 4y agoIt's true. Basically every estimation or inference problem can be solved by MCMC. The only reason not to do so is a lack of computational power.
- operator-name 4y agoThe majority of estimation problems have either: - too much uncertainty - extremely relaxed tolerances for acceptable answers and are performed in people's heads or simple napkin/grade school maths. It's not about computational power, it's about what's right for the job.