Re: AI-GEOSTATS: difference between Monte Carlo and bootstrap
> Hi everyone,In bootstrapping you create samples which have not been observed directly
> I would appreciate any light in defining and separating Monte Carlo
> resampling techniques and bootstrapping techniques. In my search
> it seems that the two notions are more or less the same .... which
> puzzles me - i suppose if there are 2 different names, would be 2
> different definitions - but maybe i am wrong.
from samples which have been observed, while in Monte Carlo you create
samples which have not been observed indirectly from a model for samples
which have been observed. In bootstrap an observed sample is turned into a
finite population from which to re-sample, while in Monte Carlo a model
derived from an observed sample is used to produce an infinite set of
populations from which to re-sample.
There are 3 major resampling techniques: bootstrap, Monte Carlo, and
randomization (aka permutation tests). In the latter, you create samples
which have not been observed by swapping the indices for the data in
samples which have been observed instead of creating entirely new data.
One problematic question which is made transparent by resampling
techniques, but which also applies to most of statistical inference, is:
would you base your inference on data which have not been observed?
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org