Science and Statistics.¶
Box, G. E. P. (1976). Science and Statistics. Journal of the American Statistical Association, 71(356), 791-799.
Cited by¶
13 citations across 13 artifacts.
Each citation links to the sentence it supports in the citing article.
Primes¶
- Foreseeing (Prediction)
- Box's 1976 maxim
This sourceSource of the aphorism 'all models are wrong, but some are useful' (stated twice); supports the pragmatic stance that predictions are approximations whose usefulness is judged against outcomes rather than logical perfection.
- Box's 1976 maxim
- Inquiry-Change Learning Loop
- An analyst who would otherwise need a full specification of a ward, a market, or a plant before acting needs instead a model precise enough to make one falsifiable commitment.
This sourcePresents the iterative model-data loop in which a model known to be wrong earns its keep by being specific enough to produce a checkable discrepancy.
- An analyst who would otherwise need a full specification of a ward, a market, or a plant before acting needs instead a model precise enough to make one falsifiable commitment.
Domain-specific¶
Mechanisms¶
- Growth-and-Crash Stock-Flow Model
- Mathematical Model Selection
- Its failure mode is the seductive precision of a formal model: equations lend authority, and a model can be exactly wrong with great confidence when its buried assumptions don't hold.
This sourceWarns that mathematical model conclusions are conditional on simplifying assumptions known to be false and may lack practical relevance despite formal optimality.
- Its failure mode is the seductive precision of a formal model: equations lend authority, and a model can be exactly wrong with great confidence when its buried assumptions don't hold.
- Model Boundary Definition
- Its strength is keeping a model inside its competence — "all models are wrong, but some are useful," and the boundary is what says where this one is useful.
This sourceStates that all models are wrong but some are useful, and urges attention to what is importantly wrong rather than elaborating a supposedly correct model.
- Its strength is keeping a model inside its competence — "all models are wrong, but some are useful," and the boundary is what says where this one is useful.
- Model-Class Revision
- A richer class can fit the past so well it stops constraining the future — the trap behind "all models are wrong, but some are useful"
This sourceArgues that because all models are wrong, excessive elaboration cannot make one correct and economical models should be preferred.
- A richer class can fit the past so well it stops constraining the future — the trap behind "all models are wrong, but some are useful"
- Propagation Simulation
- Its central failure mode is that the model is only as good as its structure and coefficients — a simulation can be confidently, precisely wrong, and wrapping assumptions in dynamics lends them an authority they haven't earned.
This sourceWarns that mathematical models are necessarily imperfect and can be wrong even while remaining precise and useful.
- Its central failure mode is that the model is only as good as its structure and coefficients — a simulation can be confidently, precisely wrong, and wrapping assumptions in dynamics lends them an authority they haven't earned.
- Simplification Review
- Held to George Box's maxim that all models are wrong but some are useful, its whole job is to decide which wrongness is tolerable and which must be fixed.
This sourceStates that all models are wrong but some are useful.
- Held to George Box's maxim that all models are wrong but some are useful, its whole job is to decide which wrongness is tolerable and which must be fixed.
- Simplified Simulation
- The honest posture behind it is George Box's dictum that "all models are wrong, but some are useful": the reduction is justified by fitness for the decision
This sourceArgues that model reduction should be judged by practical effectiveness rather than literal realism.
- The honest posture behind it is George Box's dictum that "all models are wrong, but some are useful": the reduction is justified by fitness for the decision
- Simulation Ensemble
- Its failure mode is that a simulation ensemble gives a precise answer about the model you fed it, not about the world.
This sourceDistinguishes statistical models from the world they represent and warns that precise model results remain conditional on the model’s assumptions.
- Its failure mode is that a simulation ensemble gives a precise answer about the model you fed it, not about the world.
- Simulation Rescaling Sweep
- Its failure mode is model validity — a sweep of a wrong model yields confident, precisely-wrong thresholds, since "all models are wrong; the question is whether they are useful."
This sourceArgues that every model is an approximation and must be judged by whether it is useful for its purpose.
- Its failure mode is model validity — a sweep of a wrong model yields confident, precisely-wrong thresholds, since "all models are wrong; the question is whether they are useful."
- Stripped-Down Simulation
- This is the standing caution behind all models are wrong, but some are useful
This sourceDocuments all models are wrong, but some are useful.
- This is the standing caution behind all models are wrong, but some are useful
Verification¶
Does it exist? Confirmed. This work's DOI resolves to a registered record, which fixes its identity. That is all it fixes.
Does it back the claim? Not recorded. Neither this nor any other of the 13 citations of this work carries a recorded support check.
Support is checked per citation rather than per work — the same source can be cited soundly in one article and wrongly in another. Per-citation recording began recently, so a citation with no recorded check is a gap in the record rather than evidence it went unchecked.
See how references were verified.
Registry ID ref:cf923d3dfcdb · see in the full table