Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Rothe-Hagen identity: ∑k=0nxx+kz(x+kzk)yy+(n−k)z(y+(n−k)zn−k)=x+yx+y+nz(x+y+nzn)\sum_{k=0}^{n}\frac{x}{x+kz}\binom{x+kz}{k}\frac{y}{y+(n-k)z}\binom{y+(n-k)z}{n-k}=\frac{x+y}{x+y+nz}\binom{x+y+nz}{n}∑k=0n​x+kzx​(kx+kz​)y+(n−k)zy​(n−ky+(n−k)z​)=x+y+nzx+y​(nx+y+nz​)

Proved
NRLFormulary.rothe_hagen_identity

by LG · Sep 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

binomial-coefficientscombinatoricsnrl-plasma-formularyspecial-functions

Rothe-Hagen identity. Let x,y,zx, y, zx,y,z be complex numbers and nnn a natural number, and write (wk)=1k!∏j=0k−1(w−j)\binom{w}{k} = \frac{1}{k!}\prod_{j=0}^{k-1}(w-j)(kw​)=k!1​∏j=0k−1​(w−j) for the generalized binomial coefficient with complex upper index. Assume the identity's denominators are all non-zero, namely

  1. x+kz≠0x + kz \ne 0x+kz=0 for every 0≤k≤n0 \le k \le n0≤k≤n,
  2. y+kz≠0y + kz \ne 0y+kz=0 for every 0≤k≤n0 \le k \le n0≤k≤n,
  3. x+y+nz≠0x + y + nz \ne 0x+y+nz=0.

Then

∑k=0nxx+kz(x+kzk)  yy+(n−k)z(y+(n−k)zn−k)  =  x+yx+y+nz(x+y+nzn).\sum_{k=0}^{n} \frac{x}{x+kz}\binom{x+kz}{k}\;\frac{y}{y+(n-k)z}\binom{y+(n-k)z}{n-k} \;=\; \frac{x+y}{x+y+nz}\binom{x+y+nz}{n}.k=0∑n​x+kzx​(kx+kz​)y+(n−k)zy​(n−ky+(n−k)z​)=x+y+nzx+y​(nx+y+nz​).

This is the identity printed on p. 3 of the NRL Plasma Formulary, where it is stated to hold for all complex xxx, yyy, zzz "except when singular"; the three hypotheses above are that proviso written out. Setting z=0z = 0z=0 recovers the Vandermonde convolution, and the identity is the coefficient form of the multiplicativity of the generalized binomial series.

Formalization Note. The sum runs over k∈{0,1,…,n}k \in \{0, 1, \ldots, n\}k∈{0,1,…,n} and the complementary index is the natural-number difference n−kn - kn−k, which is the ordinary difference throughout that range.

Preamble
import Mathlib
import Definitions.Def_NRLFormulary_cbinom
Formal statement
namespace NRLFormulary
theorem rothe_hagen_identity (x y z : ℂ) (n : ℕ)
    (hx : ∀ k ≤ n, x + (k : ℂ) * z ≠ 0)
    (hy : ∀ k ≤ n, y + (k : ℂ) * z ≠ 0)
    (hxy : x + y + (n : ℂ) * z ≠ 0) :
    ∑ k ∈ Finset.range (n + 1),
        (x / (x + (k : ℂ) * z) * cbinom (x + (k : ℂ) * z) k) *
          (y / (y + ((n - k : ℕ) : ℂ) * z) * cbinom (y + ((n - k : ℕ) : ℂ) * z) (n - k))
      = (x + y) / (x + y + (n : ℂ) * z) * cbinom (x + y + (n : ℂ) * z) n := by sorry
end NRLFormulary
Source
J. D. Huba, NRL Plasma Formulary, Naval Research Laboratory, 2013, p. 3, section "Numerical and Algebraic", the Rothe-Hagen identity ("good for all complex x, y, z except when singular"), footnote 2: H. W. Gould, "Note on Some Binomial Coefficient Identities of Rosenbaum", J. Math. Phys. 10, 49 (1969); H. W. Gould and J. Kaucky, "Evaluation of a Class of Binomial Coefficient Summations", J. Comb. Theory 1, 233 (1966). https://www.nrl.navy.mil/News-Media/Publications/NRL-Plasma-Formulary/
Read-back

What the Lean code literally says, in plain math · self-authored-by-drafting-agent (non-blind)

⚠️ Non-blind read-back — not independent testimony. This read-back was written by the same agent that drafted this Lean statement, not by an independent blind auditor. Its author had the source text, the intended informal statement and the mission's intent in view while writing it, so it cannot serve as an independent faithfulness check: it is the drafter's own restatement of the drafter's own code. A reviewer should read it as self-testimony and, where independence matters, obtain a genuinely blind read-back before relying on it.

The statement quantifies over three complex numbers xxx, yyy, zzz and one natural number nnn, and assumes three hypotheses:

  1. for every natural number kkk with k≤nk \le nk≤n, the complex number x+kzx + kzx+kz is nonzero;
  2. for every natural number kkk with k≤nk \le nk≤n, the complex number y+kzy + kzy+kz is nonzero;
  3. the complex number x+y+nzx + y + nzx+y+nz is nonzero.

Under these hypotheses it asserts the equality of two complex numbers:

∑k=0n(xx+kz⋅C(x+kz, k))(yy+(n−k)z⋅C(y+(n−k)z, n−k))  =  x+yx+y+nz⋅C(x+y+nz, n),\sum_{k=0}^{n} \left(\frac{x}{x+kz}\cdot C(x+kz,\,k)\right)\left(\frac{y}{y+(n-k)z}\cdot C(y+(n-k)z,\,n-k)\right) \;=\; \frac{x+y}{x+y+nz}\cdot C(x+y+nz,\,n),k=0∑n​(x+kzx​⋅C(x+kz,k))(y+(n−k)zy​⋅C(y+(n−k)z,n−k))=x+y+nzx+y​⋅C(x+y+nz,n),

where C(w,m)C(w,m)C(w,m) abbreviates the function supplied by this bundle, namely C(w,m)=1m!∏j=0m−1(w−j)C(w,m) = \frac{1}{m!}\prod_{j=0}^{m-1}(w-j)C(w,m)=m!1​∏j=0m−1​(w−j), a product of mmm factors divided by the factorial of mmm, with C(w,0)=1C(w,0) = 1C(w,0)=1.

Details the code fixes:

  • The summation index kkk runs over the n+1n+1n+1 values 0,1,…,n0, 1, \ldots, n0,1,…,n.
  • The complementary index n−kn-kn−k is truncated natural-number subtraction. Because k≤nk \le nk≤n for every term of the sum, it coincides with the ordinary difference; the truncation at 000 is never reached.
  • All divisions are complex divisions. By hypotheses 1-3 every denominator occurring in the statement is nonzero, so Lean's convention that division by zero yields zero is not exercised.
  • The hypotheses quantify over all k≤nk \le nk≤n, hence include k=0k = 0k=0: they imply x≠0x \ne 0x=0 and y≠0y \ne 0y=0.
  • The hypotheses are satisfiable — for example x=y=1x = y = 1x=y=1, z=0z = 0z=0, any nnn — so the statement is not vacuous. At n=0n = 0n=0 it reduces to xx⋅yy=x+yx+y\frac{x}{x}\cdot\frac{y}{y} = \frac{x+y}{x+y}xx​⋅yy​=x+yx+y​.
  • Nothing else is assumed about xxx, yyy, zzz: they may be real, zero (subject to the hypotheses), or of any argument.

The statement is currently recorded with its proof left open.

Human review
  • Endorsed by Shuze Chen · Sep 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by LG · Sep 24, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me