Inversion of the Pieri formula for Macdonald polynomials (Q2496239): Difference between revisions

The ``symmetric functions \(P_\lambda(x;q,t)\),'' introduced by \textit{I. G. Macdonald} in [A new class of symmetric functions, Sémin. Lothar. Comb. 20, B20a (1988; Zbl 0962.05507)], and now better known as ``the Macdonald polynomials,'' have turned out to be of crucial importance in several areas of mathematics and mathematical physics since then [cf. \textit{I.~G.~Macdonald}, Symmetric functions and Hall polynomials, 2nd ed., Oxford Univ. Press (1995; Zbl 0824.05059) and Affine Hecke algebras and orthogonal polynomials, Cambridge University Press, Cambridge (2003; Zbl 1024.33001)]. Although formulas for these polynomials existed for some time [cf. p.~346 in \textit{I.~G.~Macdonald}, Symmetric functions and Hall polynomials (loc. cit.) and \textit{L.~Lapointe, A.~Lascoux} and \textit{J.~Morse}, Int. Math. Res. Not. 1998, No.~18, 957--978 (1998; Zbl 0916.05077)], these are not always satisfying because they involve summations over large combinatorial sets, which are not easily generated, and therefore they are not easily implementable on a computer, say. The main results of the paper under review are two explicit ``analytic'' formulas for the Macdonald polynomials in terms of multiple summations, which are straight-forward to implement on a computer. One formula provides the explicit expansion of the Macdonald polynomials in terms of elementary symmetric functions, while the other provides an explicit expansion in terms of certain ``modified'' complete symmetric functions that appear frequently in Macdonald polynomials theory. The first formula extends the earlier well-known formula for Macdonald polynomials indexed by two-rowed partitions due to \textit{N. H. Jing} and \textit{T. Józefiak} [Duke Math. J. 67, 377--385 (1992; Zbl 0772.05096)]. In the case \(q=t\), the formulas reduce to the classical Jacobi--Trudi identities for Schur functions. The proof works by starting with the Pieri formula for Macdonald polynomials, interpreting it as a triangular system of equations, expressing the Macdonald polynomials multiplied by an elementary symmetric function in terms of the Macdonald polynomials. This system of equations is explicitly inverted, that is, the authors compute the matrix which is the inverse matrix of the coefficient matrix of the above system of equations explicitly. To accomplish this, they make use of an operator method for computing inverse matrices due to J.~Hofbauer and the reviewer [Trans. Am. Math. Soc. 305, 431--465 (1988; Zbl 0653.05007)]. Reviewer's remark: An alternative proof of the main theorem has subsequently been given by the first author in [``A short proof of generalized Jacobi-Trudi expansions for Macdonald polynomials,'' preprint, \texttt{math.CO/0401032}, in: Jack, Hall-Littlewood and Macdonald Polynomials, Contemp. Math. 417, 271--280 (2006; Zbl 1133.05099)]. It requires, however, the knowledge of the formula, while in the paper under review the formula is \textit{derived}.