All models are wrong — but some are useful

7 Feb, 2014 at 10:44 | Posted in Statistics & Econometrics | 3 Comments

The good scientist must have the flexibility and courage to seek out, recognize, and exploit such errors — especially his own. In particular, using Bacon’s analogy, he must not be like Pygmalion and fall in love with his model.

Since all models are wrong the scientist cannot obtain a “correct” one by excessive elaboration. On the contrary following William of Occam he should seek an economical description of natural phenomena. Just as the ability to devise simple but evocative models is the signature of the great scientist so overelaboration and overparameterization is often the mark of mediocrity.

41n48U0skrLSince all models are wrong the scientist must be alert to what is importantly wrong. It is inappropriate to be concerned about mice when there are tigers abroad …

The statistician knows, for example, that in nature there never was a normal distribution, there never was a straight line, yet with normal and linear assumptions, known to be false, he can often derive results which match, to a useful approximation, those found in the real world.

It follows that, although rigorous derivation of logical consequences is of great importance to statistics, such derivations are necessarily encapsulated in the knowledge that premise, and hence consequence, do not describe natural truth. It follows that we cannot know that any statistical technique we develop is useful unless we use it. Major advances in science and in the science of statistics in particular, usually occur, therefore, as the result of the theory-practice iteration.

George E. P. Box (1919-2013)


  1. I think that expression “all models” have no mean and is too much extense, Godell theorem and correspondent Turing limitations in his theoric computing machines make proof of this.

  2. The difference between the idealistic disposition of rationalism stemming from Platonism, which holds that ideas are perfect and events are imperfections that don’t conform perfectly with the model, and the realistic disposition of empiricism stemming from Hume, which holds that experience is the arbiter of ideas and that formal and conceptual models only approximate experience in an ever-changing context of events that occur under different conditions. People of different disposition lean in one direction of the other, and it is the dialectic between them that leads to progress in knowledge that is able to navigate between the Scylla of dogmatism and Charybdis of skepticism.

  3. […] All models are wrong — but some are useful […]

Sorry, the comment form is closed at this time.

Blog at
Entries and Comments feeds.