How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Nonidentifiability precludes simultaneous consistency
Statement
Let be a parametric statistical model, let be a metric-valued estimand, and let be a nonrandomized estimator sequence for IID samples from this model. If is not identifiable, then there is a pair of observationally equivalent parameters with distinct target values, and no estimator sequence is consistent in probability at both parameters.
Facts & Assumptions
Given: The model, estimand, estimator sequence, and hypotheses in the statement.
Parametric statistical models: A parametric statistical model is the parameter map with .
Observational equivalence: exactly when as measures on the observation space.
Estimands on a statistical model: An estimand is a specified function .
Identifiability of an estimand: An estimand is identifiable exactly when implies .
Estimator sequences in IID experiments: The size- experiment has law on , and an estimator is a measurable map on that space.
Pointwise consistency in probability: Consistency at means for every .
The product sigma-algebra and its finite iterates: The finite product sigma-algebra is generated recursively by measurable rectangles.
For sigma-finite factors, the product measure exists, has the rectangle formula, is sigma-finite, and is unique: The product measure exists and gives each measurable rectangle the product of the factor measures.
Measures agreeing on a generating pi-system are equal under an increasing finite-measure exhaustion from that pi-system: Measures agreeing on a generating -system and on an increasing finite-measure exhaustion from it agree on the generated sigma-algebra.
Metric space: iff , symmetry, and the triangle inequality; pseudometric and ultrametric: A metric is nonnegative, symmetric, separates distinct points, and satisfies the triangle inequality.
A measurable function between measurable spaces: A map is measurable when the preimage of every target-measurable set is measurable.
The Borel sigma-algebra of a topological space: The Borel sigma-algebra of a metric space contains every closed subset.
Arbitrary unions and finite intersections of open sets are open, open balls are open and closed balls are closed: Every closed ball in a metric space is closed.
Proof
Given: The objects and hypotheses in the statement.
By [F4], nonidentifiability supplies with and . Fix this pair.
Fix . By [F1], the observation laws in the model are . Measurable rectangles in form a -system generating by [F7]. For every rectangle , [F8] and [F2] give . The full space is itself a rectangle and is an increasing exhaustion of finite mass one. Applying [F9] shows on all of .
Put and . The two closed target balls are disjoint: a point in both would give by [F10].
Suppose, for contradiction, that the estimator sequence is consistent at both parameters. For let . These events are measurable because each is closed by [F13] and Borel by [F12], while is measurable by [F5] and preimages of Borel sets are measurable by [F11]. By [F6], for each .
For each , step 2.1 gives one common sample law for both probabilities, and step 2.2 gives disjoint events, so their probabilities sum to at most one. Step 3.1 says both probabilities tend to one, so their sum tends to two, a contradiction. Therefore the estimator sequence cannot be consistent at both parameters.
Depends on
- The Borel sigma-algebra of a topological space
- Pointwise consistency in probability
- Estimands on a statistical model
- Estimator sequences in IID experiments
- Identifiability of an estimand
- A measurable function between measurable spaces
- Metric space: $d(x,y) = 0$ iff $x = y$, symmetry, and the triangle inequality; pseudometric and ultrametric
- Observational equivalence
- Parametric statistical models
- The product sigma-algebra and its finite iterates
- Statistical experiments
- Measures agreeing on a generating pi-system are equal under an increasing finite-measure exhaustion from that pi-system
- Arbitrary unions and finite intersections of open sets are open, open balls are open and closed balls are closed
- For sigma-finite factors, the product measure exists, has the rectangle formula, is sigma-finite, and is unique
Used by
Dependency tree · two levels
44 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.
Sources
- John Duchi, Theory of Statistics, Stats 300B Lecture 3, Winter 2018, PDF p. 4 (standard reference, not scraped)
- Peter Bartlett, Theoretical Statistics, Stat 210B Lecture 15, PDF pp. 6–8 (standard reference, not scraped)