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.
Influence function and sensitivity of the median
Statement
Let be a Borel probability law on , let be its lower median, and assume . Suppose that on an open interval containing there is a continuous density version satisfying
For each fixed , the right point-contamination influence function of the lower-median functional is
Consequently, its gross-error sensitivity is .
Facts & Assumptions
Given: A Borel probability law , its lower median , the equality , and a continuous local density on an interval with . Fix a contamination point .
The lower median of a real probability law: on its stated domain.
Statistical functionals and gross-error influence functions: is the right derivative of at , when finite.
Gross-error sensitivity of a functional: is the supremum of the absolute influence values over the directions where they exist.
Proof
Since is open, is continuous at , and , choose and so that and throughout this closed interval. For in it, the density identity gives . Thus is continuous and strictly increasing there, with .
Suppose first that , and shrink so that . For sufficiently small , the number lies strictly between and . Continuity and strict increase give a unique with . Since as , for every eventually , so and . Since , the mixture cdf at is . For , the contamination term is zero and, by monotonicity and , the mixture cdf is . At , monotonicity gives , so . For , the contamination term is and : below the local interval this follows from monotonicity and , and inside it from strict increase. Therefore the mixture cdf at such is . Together with the cases and , this puts the mixture cdf below for every , while it equals at , so [F2] gives .
If , shrink so that . For small , lies strictly between and . There is a unique with . Since as , for every eventually , so and . Now , so the mixture cdf at is . For every , the contamination term is zero; if , strict increase on the local interval gives , and if then monotonicity gives . Thus the mixture cdf is below for every , while it equals at , so [F2] gives .
In the case , step 2.1 gives and . The density identity and continuity of imply . Dividing these two relations yields .
In the case , step 2.2 gives and . The same density quotient tends to , hence .
If , continuity of at gives and . Monotonicity then puts every below the median threshold, while itself is above it, so the lower median remains exactly and its right derivative is . By [F3] and steps 3.1–3.2, these three cases give the displayed pointwise influence function.
The absolute influence value is for every and zero at . Both sides of are nonempty in , so the supremum over all directions is by [F4]. Since has a density near , ; therefore the conventional expression agrees with this pointwise derivative -almost everywhere, though its value at is not the literal right derivative.
Remarks
The conventional indicator formula in the Statement of the controlling design is an -almost-everywhere representative. At the single point , the right point-contamination derivative of the lower median is ; retaining this pointwise value makes the definition agree exactly with the computed limit.
Depends on
Used by
Dependency tree · two levels
11 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
- P. B. Stark, Statistics 240 Lecture Notes, Part 9: Robustness and Related Topics, §§1.2–1.3, PDF pp. 2–3, 13–18 (standard reference, not scraped)
- Elvezio Ronchetti, Introduction to Robust Statistics, slides 15–23 (standard reference, not scraped)
- Pavlo Mozharovskyi, Introduction to Robust Statistics, slide ‘Sample median: influence function’ (PDF p. 32) (standard reference, not scraped)