3 ms·
More generally: All models are wrong, but some are useful. -- George E. P. Box, in "Science and Statistics" (1976) <https://www-sop.inria.fr/members/Ian.Jermy
by dredmorbius 3y ago
More generally: All models are wrong, but some are useful.
-- George E. P. Box, in "Science and Statistics" (1976) <https://www-sop.inria.fr/members/Ian.Jermyn/philosophy/writings/Boxonmaths.pdf https://www-sop.inria.fr/members/Ian.Jermyn/philosophy/writi...> (PDF)
See: <https://en.wikipedia.org/wiki/All_models_are_wrong https://en.wikipedia.org/wiki/All_models_are_wrong>
Box apparently expanded on the notion in Box, G. E. P. (1979), "Robustness in the strategy of scientific model building", in Launer, R. L.; Wilkinson, G. N. (eds.), Robustness in Statistics, Academic Press, pp. 201–236, doi:10.1016/B978-0-12-438150-6.50018-2, ISBN 9781483263366.
Full text online at <https://pure.au.dk/ws/files/224156357/skogen_etal_2021_meps.pdf https://pure.au.dk/ws/files/224156357/skogen_etal_2021_meps....> (PDF)