Papers
Topics
Authors
Recent
Search
2000 character limit reached

Raney Extension: Combinatorics and Beyond

Updated 9 July 2026
  • Raney extension is a family of mathematically distinct generalizations that replace one-parameter structures with richer two-parameter objects, preserving key recurrences and dualities.
  • Various combinatorial models—such as p-stars, coral diagrams, threshold sequences, and colored Dyck paths—illustrate concrete extensions of Catalan and Fuss–Catalan numbers.
  • Analytic and algebraic extensions, including Raney distributions and Raney–Mohanty identities, bridge random matrix theory, equilibrium problems, and pointfree topology through rigorous combinatorial methods.

In current usage across the cited literature, the expression Raney extension does not denote a single construction. It names a family of mathematically distinct extensions inspired by ideas of G. N. Raney: a two-parameter extension of Catalan and Fuss–Catalan combinatorics through the Raney numbers; a multivariable extension of Rothe- and Gould-type convolution identities; a pointfree extension of frame-based duality by adjoining a coframe of saturated-type objects to a frame of opens; and an automata-theoretic extension of continued-fraction transduction used to study generalized Lagrange spectra (Zhou, 2015, Guo, 2010, Suarez, 2024, Dong et al., 2024). The common theme is the replacement of a one-parameter or one-level structure by a richer two-level or two-parameter object that preserves a recognizable Raney-type recurrence, transform, or duality.

1. Raney numbers as a two-parameter Catalan extension

The basic combinatorial form of a Raney extension is the passage from Catalan numbers to the Raney numbers

Rp,r(k)=rkp+r(kp+rk),R_{p,r}(k)=\frac{r}{kp+r}\binom{kp+r}{k},

where pp is a positive integer and r,kr,k are nonnegative integers. The Catalan numbers occur as R2,1(k)R_{2,1}(k), and the Fuss–Catalan numbers occur as Rp,1(k)R_{p,1}(k). Hilton–Pedersen’s convolution formula

Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)

shows that Rp,r(k)R_{p,r}(k) is an rr-fold convolution of the Fuss–Catalan sequence, so the second parameter rr extends the usual Catalan/Fuss–Catalan arity parameter by introducing a nontrivial base multiplicity (Zhou, 2015).

This extension is reflected at the level of recurrences. For r=1r=1,

pp0

and for pp1,

pp2

with pp3 and pp4 for pp5. When pp6, the first recurrence reduces to the classical Catalan recurrence, so the paper explicitly realizes a literal extension of Catalan recursion to a two-parameter Raney family. The proof uses pp7-stars and coral diagrams: a pp8-star is a rooted tree with one base vertex and pp9 terminal edges above it, and a coral diagram of type r,kr,k0 starts from an r,kr,k1-star and iteratively attaches r,kr,k2 r,kr,k3-stars tier by tier (Zhou, 2015).

The same paper connects this Raney extension to simultaneous core partitions. If r,kr,k4 with r,kr,k5, then the number of r,kr,k6-core partitions whose parts are multiples of r,kr,k7 is

r,kr,k8

The proof passes through Anderson’s poset r,kr,k9, the R2,1(k)R_{2,1}(k)0-set description of core partitions, and the property R2,1(k)R_{2,1}(k)1 that every maximal run of consecutive integers has length divisible by R2,1(k)R_{2,1}(k)2. The count of order ideals of R2,1(k)R_{2,1}(k)3 with property R2,1(k)R_{2,1}(k)4 satisfies the same recurrences and initial conditions as R2,1(k)R_{2,1}(k)5, confirming Amdeberhan’s conjecture and giving a new combinatorial interpretation of these Raney numbers (Zhou, 2015).

2. Alternative combinatorial models

Several later papers broaden the combinatorial meaning of Raney numbers without altering the closed form. One direction uses threshold sequences. For integers R2,1(k)R_{2,1}(k)6, R2,1(k)R_{2,1}(k)7, and R2,1(k)R_{2,1}(k)8, a R2,1(k)R_{2,1}(k)9-threshold sequence of length Rp,1(k)R_{p,1}(k)0 is a strictly increasing sequence Rp,1(k)R_{p,1}(k)1 with Rp,1(k)R_{p,1}(k)2. These sequences are in bijection with ordered Rp,1(k)R_{p,1}(k)3-tuples of Rp,1(k)R_{p,1}(k)4-ary trees with total Rp,1(k)R_{p,1}(k)5 internal nodes, so they are counted by

Rp,1(k)R_{p,1}(k)6

The same paper also identifies proper threshold sequences with certain Motzkin-like paths having long up and down steps, producing path models for the same Raney numbers (Rusu, 2021).

A second direction uses planar embeddings. Coral diagrams can be reinterpreted as planar embeddings of trees with Rp,1(k)R_{p,1}(k)7 internal vertices such that all internal vertices are Rp,1(k)R_{p,1}(k)8-valent except the vertex incident upon the leftmost terminal edge, which is Rp,1(k)R_{p,1}(k)9-valent. This yields an ordered-partition identity

Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)0

where the sum is over ordered partitions Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)1 of Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)2. The same framework gives specific interpretations such as Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)3 as oriented trees with the source or sink property, and it identifies Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)4 with connected non-elliptic Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)5 webs that lack an internal face and have a constant boundary string with Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)6 pluses (Beagley et al., 2015).

A third direction uses Dyck paths on colored lattices. For column coloring modulo Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)7, a Dyck path of semilength Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)8 is called Rp,r(k)=i1++ir=kCp(i1)Cp(ir)R_{p,r}(k)=\sum_{i_1+\cdots+i_r=k} C_p(i_1)\cdots C_p(i_r)9-balanced if each residue class of columns carries the same number of cells below the path. Writing Rp,r(k)R_{p,r}(k)0 and Rp,r(k)R_{p,r}(k)1, the number of such paths is

Rp,r(k)R_{p,r}(k)2

The proof uses a structural lemma forcing the path to be constant on blocks of Rp,r(k)R_{p,r}(k)3 consecutive columns and then applies the cycle lemma to convert the problem to generalized ballot sequences. For Rp,r(k)R_{p,r}(k)4, this recovers Fried’s column-alternating enumeration; for Rp,r(k)R_{p,r}(k)5, it places further Catalan-type sequences into the Raney family (Saikia, 8 Jul 2026).

Taken together, these models suggest that the Raney extension of Catalan combinatorics is not tied to a single avatar. Trees, core partitions, threshold sequences, generalized Motzkin paths, modularly colored Dyck paths, and web diagrams all realize the same two-parameter counting law.

3. Analytic and probabilistic extensions: Raney distributions

A second major use of Raney extension is analytic. For Rp,r(k)R_{p,r}(k)6 and Rp,r(k)R_{p,r}(k)7, the sequence

Rp,r(k)R_{p,r}(k)8

is the moment sequence of a probability measure Rp,r(k)R_{p,r}(k)9 on rr0, called the Raney distribution. For rational rr1, the measure is absolutely continuous with density rr2, supported on

rr3

and expressible in terms of Meijer rr4-functions and generalized hypergeometric functions. Special cases include the Marchenko–Pastur distribution rr5, the Fuss–Catalan family rr6, and the semicircle law centered at rr7, which appears as rr8. The identity

rr9

exhibits the diagonal line rr0 as a one-parameter generalization of the semicircle law (Mlotkowski et al., 2012).

This probabilistic extension is closely tied to random matrix theory. For a product of rr1 square random Ginibre matrices, the limiting squared singular-value density is the Fuss–Catalan distribution rr2. The same paper then develops exact expressions for the full two-parameter family rr3, interpreting them as a two-parameter generalization of both Fuss–Catalan distributions and the Wigner semicircle law. In this sense, the Raney extension enlarges the known spectral laws of product ensembles from the rr4 line to the full admissible region rr5 (Penson et al., 2011).

A further extension appears in equilibrium problems. For rr6, rr7, and more generally rr8, explicit logarithmic energy functionals are shown to have equilibrium densities whose moments are the corresponding Raney numbers. Two methods are used: Wiener–Hopf factorization of the singular integral equation and analysis of the algebraic equation satisfied by the Green’s function. This extends previously conjectured equilibrium descriptions from special integer cases to general rr9 and integer r=1r=10 (Forrester et al., 2014).

The same algebraic viewpoint also extends beyond compactly supported Raney laws. The Stieltjes transform of the limiting spectral density for products formed from r=1r=11 inverse standard Gaussian matrices and r=1r=12 standard Gaussian matrices satisfies a variant of the Raney algebraic equation and admits a simple parameterization on r=1r=13. The leading asymptotics at the endpoints show universal features analogous to those of the compactly supported Raney distributions (Forrester et al., 2014).

4. Raney–Mohanty identities and multinomial convolution

In another strand of the literature, Raney extension refers to a multivariable extension of classical convolution identities. Starting from Rothe’s identity

r=1r=14

Mohanty’s multivariable generalization replaces the scalar shift r=1r=15 by a dot product r=1r=16 and ordinary binomial coefficients by multinomial-type coefficients. The first Raney–Mohanty identity is

r=1r=17

and the second is

r=1r=18

In the terminology of the paper, these are Raney–Mohanty identities because Raney had already obtained important special cases in work on Lagrange inversion (Guo, 2010).

The proof strategy is purely combinatorial. The paper uses a graded alphabet r=1r=19 with pp00 and pp01, then counts words of fixed length, weight, and composition in two ways. The coefficient

pp02

is interpreted as the number of words with prescribed counts of heavy letters. Bijective manipulations of prefixes and suffixes then yield bijective proofs of both Gould–Mohanty and Raney–Mohanty identities. In this setting, a Raney extension is the passage from one-variable binomial convolution to multivariable multinomial convolution, with the same global “sum equals one coefficient” structure preserved (Guo, 2010).

5. Pointfree and algebraic Raney extensions

A completely different meaning of the term appears in pointfree topology. Here a Raney extension is a pair consisting of a frame and a coframe. In the notation of the pointfree/topological papers, it is written pp03, where pp04 is a frame, pp05 is a coframe, pp06 is a subframe, pp07 meet-generates pp08, and the inclusion preserves strongly exact meets. The motivating example is

pp09

where pp10 is the frame of opens of a space pp11 and pp12 is the coframe of saturated sets. The paper "Raney extensions: a pointfree theory of pp13 spaces based on canonical extension" proves a dual adjunction between pp14 and pp15, with all pp16 spaces as fixpoints. It also places every Raney extension between two canonical constructions over a frame pp17,

pp18

whose spectra are respectively the classical spectrum pp19 and the pp20 spectrum pp21 (Suarez, 2024).

The companion paper on topological aspects develops the categorical consequences. It characterizes sobriety, the pp22 axiom, and the pp23 axiom for spaces in terms of algebraic properties of their Raney duals; defines sobriety, pp24, and pp25 for general Raney extensions; proves that a sober coreflection always exists; proves that a pp26 reflection exists when morphisms are restricted to exact maps; and shows that a frame is subfit if and only if it admits a pp27 Raney extension, while a subfit frame is scattered if and only if it admits a unique Raney extension (Suarez, 2024).

The algebraic culmination is the MT-algebra paper, which reverses the order of notation and writes a Raney extension as pp28. There pp29 is a coframe, pp30 is a subframe that meet-generates it, and the mixed distributivity law

pp31

is imposed for pp32 and pp33. The paper introduces Raney morphisms between McKinsey–Tarski algebras, generalizes the Funayama envelope construction, proves that the resulting category is equivalent to the category of Raney extensions, and defines the pp34-hull of a Raney extension as a generalization of the pp35-hull of a frame (Bezhanishvili et al., 1 Sep 2025). This makes the pointfree Raney extension a categorical bridge between frame theory, coframe completions, and interior-algebraic structures.

6. Automata, sequences, and generalized Raney trees

Raney’s name also appears in discrete algorithmic settings where the emphasis is on extension procedures rather than on two-parameter counting formulas. For a prime pp36, the paper on Raney transducers studies the pp37-Lagrange spectrum

pp38

through finite-state transducers acting on LR-expansions of continued fractions. It defines slow and fast Raney transducers, gives an algorithm that computes pp39 if it terminates, conjectures that it always terminates, and verifies for primes pp40 that pp41 is the square root of a rational number. The same computations reveal that the highest values of pp42 occur for the Heegner primes pp43, pp44, and pp45, and that the continued fractions of pp46 and pp47 realizing the minimum fall into one of three symmetric relations (Dong et al., 2024).

A different extension problem appears for binomid indices, equivalently Raney sequences. A finite binomid index pp48 has a unique lexicographically minimal infinite binomid extension pp49, and this extension is necessarily eventually periodic. The paper provides a formula for the minimal period,

pp50

an upper bound on the preperiod, and a monoid-theoretic description: the monoid of lex-minimal extensions is an inductive limit of finitely presented monoids (Caalim et al., 26 May 2025). Here the Raney extension is literally an extension of finite combinatorial data to a canonical infinite eventually periodic object.

The orbit-counting paper on symmetric discrete interval exchanges gives yet another discrete generalization. It defines a recursive function pp51 on compositions pp52 that counts the number of orbits of the associated symmetric discrete interval exchange, shows that minimal exchanges reduce to the composition pp53 labeling the root of the classical Raney tree, and constructs a generalized tree of circular compositions in which every circular composition appears exactly once (Lapointe, 2018). In this setting, the Raney extension is a tree-theoretic enlargement of the classical Raney tree from coprime pairs to arbitrary circular compositions.

Across these discrete settings, the shared pattern is canonical prolongation: finite words, finite indices, or finite compositions are pushed through a Raney-type automaton or recursion until periodic, minimal, or tree-like structure emerges.

Topic to Video (Beta)

No one has generated a video about this topic yet.

Whiteboard

No one has generated a whiteboard explanation for this topic yet.

Follow Topic

Get notified by email when new papers are published related to Raney Extension.