relative_lorenz_curve_zero
For any natural number n, population vector x in stdSimplex ℝ (Fin n), and strictly positive Gibbs vector γ in stdSimplex ℝ (Fin n), the relative Lorenz curve relative_lorenz_curve x γ 0 equals 0.
Statement
theorem relative_lorenz_curve_zero {n : ℕ} {x γ : Fin n → ℝ}
(hx : x ∈ stdSimplex ℝ (Fin n)) (hγ : γ ∈ stdSimplex ℝ (Fin n)) (hγ_pos : ∀ i, 0 < γ i) :
relative_lorenz_curve x γ 0 = 0- Source
- arXiv:2610.35433:874
- Verified
- 01 Nov 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 1
- relative_lorenz_curveGiven a natural number n, vectors x, γ : Fin n → ℝ, and a parameter t : ℝ, the relative Lorenz curve relative_lorenz_curve x γ t is…
Read back from the Lean
For any natural number n and any vectors x, γ : Fin n → ℝ, if x lies in the standard simplex stdSimplex ℝ (Fin n) (meaning x i ≥ 0 for all i and ∑ i, x i = 1), γ lies in the standard simplex stdSimplex ℝ (Fin n) (meaning γ i ≥ 0 for all i and ∑ i, γ i = 1), and every component of γ is strictly positive (0 < γ i for all i), then relative_lorenz_curve x γ 0 = 0.
Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.
The proof, explained
The theorem shows that the relative Lorenz curve evaluates to 0 at t = 0. Unfolding the definition reveals that the curve value is the supremum of ∑ i, α i * x i over all weight vectors α satisfying 0 ≤ α i ≤ 1 and ∑ i, α i * γ i ≤ 0. Because the Gibbs vector γ has strictly positive components and α is nonnegative, each product α i * γ i is nonnegative; since their sum is at most 0, each α i must be 0, so the feasible set contains only the zero function. The image of the objective function on this singleton is {0}, whose supremum is 0.
By definition, the relative Lorenz curve at t = 0 is the supremum of ∑ i, α i * x i over all weight vectors α satisfying 0 ≤ α i ≤ 1 for all i and ∑ i, α i * γ i ≤ 0.
We unfold the definition of relative_lorenz_curve evaluated at parameter t = 0.
unfold relative_lorenz_curveThe set of weight vectors α satisfying 0 ≤ α i ≤ 1 for all i and ∑ i, α i * γ i ≤ 0 is precisely the singleton {0}.
We show set extensionality by establishing both inclusions. For the forward direction, suppose α satisfies the constraints 0 ≤ α i ≤ 1 and ∑ i, α i * γ i ≤ 0. Because each γ i > 0, every product α j * γ j is nonnegative (mul_nonneg, le_of_lt). Since each term in a sum of nonnegative terms is bounded by the total sum (Finset.single_le_sum, Finset.mem_univ), we have 0 ≤ α i * γ i ≤ ∑ j, α j * γ j ≤ 0, which forces α i * γ i = 0 by linear arithmetic. By mul_eq_zero.mp, either α i = 0 or γ i = 0; but γ i > 0 implies γ i ≠ 0 (ne_of_gt), so α i = 0 for every index i, meaning α = 0. For the reverse direction, the zero function trivially satisfies 0 ≤ 0 ≤ 1 and ∑ i, 0 * γ i = 0 ≤ 0.
uses
Set.mem_singleton_iff,mul_nonneg,le_of_lt,Finset.single_le_sum,Finset.mem_univ,mul_eq_zero.mp,ne_of_gthave h_set : {α : Fin n → ℝ | (∀ i, 0 ≤ α i ∧ α i ≤ 1) ∧ ∑ i, α i * γ i ≤ 0} = {(0 : Fin n → ℝ)} := by ext α simp only [Set.mem_singleton_iff] constructor · rintro ⟨h1, h2⟩ ext i have h_nonneg : ∀ j, 0 ≤ α j * γ j := fun j => mul_nonneg (h1 j).1 (le_of_lt (hγ_pos j)) have : α i * γ i ≤ ∑ j, α j * γ j := Finset.single_le_sum (fun j _ => h_nonneg j) (Finset.mem_univ i) have h_i_nonneg := h_nonneg i have h_term : α i * γ i = 0 := by linarith cases mul_eq_zero.mp h_term with | inl ha => exact ha | inr hg => exact (ne_of_gt (hγ_pos i) hg).elim · rintro rfl simp
The supremum of the objective function over the singleton {0} evaluates to 0, completing the proof.
Rewriting the feasible set with the singleton {0}, the image under α ↦ ∑ i, α i * x i becomes the singleton {∑ i, 0(i) * x i} (Set.image_singleton). Simplifying the values of the zero function (Pi.zero_apply), the products with zero (zero_mul), and the sum of zeros (Finset.sum_const_zero) reduces the image to {0}. Finally, the supremum of {0} is 0 by csSup_singleton.
uses
Set.image_singleton,Pi.zero_apply,zero_mul,Finset.sum_const_zero,csSup_singletonrw [h_set, Set.image_singleton] simp only [Pi.zero_apply, zero_mul, Finset.sum_const_zero] exact csSup_singleton 0
csSup_singleton- The supremum of a singleton set {a} is a.
Finset.mem_univ- Every element of a finite type belongs to its universal finset.
Finset.single_le_sum- An element in a finite set is bounded above by the sum of all elements when all terms are nonnegative.
Finset.sum_const_zero- The sum of zeros over any finite set is zero.
le_of_lt- A strict inequality a < b implies the non-strict inequality a ≤ b.
mul_eq_zero.mp- If the product of two real numbers is zero, at least one of the factors is zero.
mul_nonneg- The product of two nonnegative numbers is nonnegative.
ne_of_gt- If b < a, then a ≠ b.
Pi.zero_apply- The zero function evaluated at any input returns zero.
Set.image_singleton- The image of a singleton set {a} under a function f is the singleton {f(a)}.
Set.mem_singleton_iff- An element x belongs to the singleton set {b} if and only if x = b.
zero_mul- Zero multiplied by any number equals zero.
Written by a model (gemini-3.8-flash) from the Lean proof and the goal states Lean reported at each step. Lean verified the proof; this explanation is a reading aid and is not itself machine-checked.
Lean source view module on GitHub
Choose a tactic — a marked line — to see the goals Lean had before and after it, as in Lean’s infoview. Step with ← → or j k.
/-- The relative Lorenz curve evaluates to 0 at t = 0. -/theorem relative_lorenz_curve_zero {n : ℕ} {x γ : Fin n → ℝ} (hx : x ∈ stdSimplex ℝ (Fin n)) (hγ : γ ∈ stdSimplex ℝ (Fin n)) (hγ_pos : ∀ i, 0 < γ i) : relative_lorenz_curve x γ 0 = 0 := by unfold relative_lorenz_curve have h_set : {α : Fin n → ℝ | (∀ i, 0 ≤ α i ∧ α i ≤ 1) ∧ ∑ i, α i * γ i ≤ 0} = {(0 : Fin n → ℝ)} := by ext α simp only [Set.mem_singleton_iff] constructor · rintro ⟨h1, h2⟩ ext i have h_nonneg : ∀ j, 0 ≤ α j * γ j := fun j => mul_nonneg (h1 j).1 (le_of_lt (hγ_pos j)) have : α i * γ i ≤ ∑ j, α j * γ j := Finset.single_le_sum (fun j _ => h_nonneg j) (Finset.mem_univ i) have h_i_nonneg := h_nonneg i have h_term : α i * γ i = 0 := by linarith cases mul_eq_zero.mp h_term with | inl ha => exact ha | inr hg => exact (ne_of_gt (hγ_pos i) hg).elim · rintro rfl simp rw [h_set, Set.image_singleton] simp only [Pi.zero_apply, zero_mul, Finset.sum_const_zero] exact csSup_singleton 0