Developments from Programming the Partition Method for a Power Series Expansion

Abstract

Recently, a novel method based on coding partitions [1]-[4] has been used to derive power series expansions to previously intractable problems. In this method the coefficients at $k$ are determined by summing the contributions made by each partition whose elements sum to $k$. These contributions are found by assigning values to each element and multiplying by an appropriate multinomial factor. This work presents a theoretical framework for the partition method for a power series expansion. To overcome the complexity due to the contributions, a programming methodology is created allowing more general problems to be studied than envisaged originally. The methodology uses the bi-variate recursive central partition (BRCP) algorithm, which is based on a tree-diagram approach to scanning partitions. Its main advantage is that partitions are generated in the multiplicity representation. During the development of the theoretical framework, scanning over partitions was seen as a discrete operation with an operator $L_{P,k}[ \cdot]$, whose summand depends on the coefficients of the two series when the original function is written as a pseudo-composite function. Simple modifications result in programs for other operators of specific types of partitions such as: (1) only odd or even elements, (2) a fixed number of elements, (3) discrete elements, (4) specific elements and (5) those restricted by element size. Another modification generates conjugate partitions by transposing Ferrers diagrams. The operator approach is then applied to the generating functions for both discrete and standard partitions. The main generalisation introduces a parameter $ω$, whose powers give the number of elements in the partitions while the coefficients become polynomials in $ω$. Finally, power series expansions for more advanced infinite products are derived, culminating in Heine's multi-parameter product.