New Binomial Identities for Fibonacci, Lucas, and Generalized Fibonacci Sequences with Multiple Indices
Published 12 Mar 2026 in math.CO | (2603.12150v1)
Abstract: This paper presents new identities expressing the terms of Fibonacci, Lucas, and generalized Fibonacci sequences with multiple indices through powers of Lucas numbers and binomial coefficients. The obtained formulas rely on the application of symmetric polynomials (Waring's formulas) to the classical Binet's formula. Particular attention is given to the binomial expansion for the generalized Fibonacci sequence, which structurally combines two adjacent binomial coefficients from Pascal's triangle.
The paper derives explicit binomial-polynomial formulas for F_nm and L_nm in terms of L_n, with parity controlled by the factor (-1)^{i(n+1)}.
The paper extends the method to arbitrary generalized Fibonacci sequences, expressing G_nm through adjacent Pascal-triangle coefficients tied to G_n and G_0.
The proofs combine Binet’s formula, symmetric-polynomial identities, Chebyshev structure, and d’Ocagne’s identity, offering exact formulas but no benchmark against fast-doubling algorithms.
How do the binomial identities for F_nm and L_nm relate to Chebyshev polynomials and Lucas triangle coefficients?
Can the generalized Fibonacci identity be proved combinatorially without using Binet’s formula or d’Ocagne’s identity?
Do these multiple-index identities extend to negative indices, higher-order recurrences, or q-Fibonacci sequences?
How might the formulas improve or compare with fast doubling and matrix exponentiation for computing large-index terms?
Find recent papers about binomial identities for generalized Fibonacci sequences.
Overview
This paper establishes three explicit binomial identities expressing multiple-index terms of the Fibonacci sequence Fnm, the Lucas sequence Lnm, and the generalized Fibonacci sequence Gnm as polynomials in the single Lucas number Ln with binomial coefficients drawn from Pascal's triangle. The proofs are elementary and uniform in method: Binet's formula reduces each multiple-index term to a symmetric function of the n-th powers of the golden-ratio conjugates α and β, and Waring's formulas for power sums convert those symmetric functions into polynomials in the elementary symmetric polynomials S=x+y=Ln and P=xy=(αβ)n=(−1)n.
Methodological framework
The paper rests on three classical results. First, Binet's formula gives Fn=(αn−βn)/5 and Lnm0, together with the relation Lnm1. Second, Waring's formulas express the two fundamental symmetric functions of two variables in terms of Lnm2 and Lnm3: the divided difference Lnm4 as an alternating sum over Lnm5, and the power sum Lnm6 as an alternating sum over Lnm7. These formulas are precisely the closed forms of Chebyshev polynomials of the second and first kinds, which the author highlights as the structural bridge between the recurrence sequences and symmetric-polynomial theory. Third, d'Ocagne's identity Lnm8 is needed only for the generalized case.
The Fibonacci and Lucas identities
Theorem on Lnm9: substituting Gnm0, Gnm1 into the divided-difference form of Waring's formula yields
Gnm2
The sign factor Gnm3 arises from combining the alternating sign of Waring's formula with the product term Gnm4. This identity expresses the ratio Gnm5 entirely as a polynomial in Gnm6, so no division or irrational arithmetic is required once Gnm7 is known — a point of direct algorithmic relevance.
Theorem on Gnm8: the power-sum form of Waring's formula gives
Gnm9
Here the coefficients Ln0 are the Lucas triangle coefficients, and the same sign factor Ln1 governs the parity structure inherited from Ln2. Both identities are direct consequences of the substitution and require no additional machinery; their correctness is fully determined by the two Waring identities, which the paper cites rather than re-proves.
The generalized Fibonacci identity
The main technical contribution is the identity for the generalized sequence Ln3 with arbitrary initial conditions Ln4. The proof exploits the representation Ln5 and eliminates Ln6 via Ln7, producing
Ln8
The Ln9-coefficient is exactly the Fibonacci ratio from the first theorem. The n0-coefficient requires the d'Ocagne step: n1, after which the Fibonacci theorem is applied with parameter n2 and the summation index is shifted by one (n3). The structural outcome is that the two sums use adjacent binomial coefficients from Pascal's triangle — n4 and n5 at matching values of n6 — with the n7 sum offset by one in its lower limit. This adjacency is presented by the author as the notable feature of the generalized formula, and it means both coefficients can be read off a single row-diagonal structure of Pascal's triangle.
One caveat in the derivation: the elimination of n8 involves division by n9, which is legitimate since α0 guarantees α1, but the intermediate expressions are rational in α2 even though the final identity is polynomial in α3 and integer-valued. The paper does not discuss whether the identity extends to negative α4 or to the Lucas companion of the generalized sequence.
Limitations and open questions
The results are confined to the two-step recurrence with characteristic roots α5; no extension to higher-order recurrences or to α6-Fibonacci analogues is attempted. The identities are proven for positive integers α7 only. The paper also does not quantify the computational advantage of the binomial form over fast doubling or matrix-power methods for evaluating α8, so the claimed algorithmic applicability remains asserted rather than benchmarked. Whether the adjacent-coefficient structure of the generalized identity admits a direct combinatorial proof — bypassing the Binet/Waring route — is left unaddressed.
Conclusion
The paper provides short, rigorous proofs that α9, β0, and β1 are polynomials in β2 with explicit binomial coefficients, unified through the application of Waring's formulas to Binet's formula. The generalized Fibonacci identity, with its paired adjacent binomial coefficients and single sign factor β3, is the paper's most distinctive result, connecting multiple-index evaluation of second-order recurrences to the combinatorics of Pascal's triangle and the Chebyshev polynomials.