I have some questions regarding Jack polynomials. I use the notation of of I.G. Macdonald's book "Symmetric Functions and Hall polynomials".

Let $\Lambda$ be the ring of symmetric functions over $\mathbb{C}$ (in countably infinitely many variables $x_i$). For a partition $\mu=(\mu_1,\mu_2,\dots)$ let $p_\mu$ denote the power sums, i.e., $p_\mu=p_{\mu_1}p_{\mu_2}\cdots$ and $p_{\mu_i}=\sum_k x_k^{\mu_i}$.

Let $\langle\ ,\ \rangle^{\alpha}$ be the inner product given by $\langle p_\mu,p_\nu\rangle=\delta_{\mu,\nu}\alpha^{\ell(\mu)}z_\mu$, where $\alpha\in\mathbb{C}$, $\ell(\mu)$ is the length of the partition $\mu$, $z_\mu=\prod_{i\in\mu}i^{m_i(\mu)}\cdot m_i(\mu)!$ and $m_i(\mu)$ is the multiplicity of $i$ in $\mu$.

The Jack symmetric functions $P_\mu^{(\alpha)}$ form a basis of $\Lambda$ indexed by partitions and are characterised by 2 properties:

  1. $P_\mu^{(\alpha)}=m_\mu + \sum_\nu u_{\mu,\nu}(\alpha)m_\nu$, where the $m_\nu$ are symmetric monomials, the $u_\nu(\alpha)\in\mathbb{C}$ and the summation index runs over all partitions $\nu <\mu$ in the dominance ordering.
  2. The Jack symmetric functions are mutually orthogonal $\langle P_\mu^{(\alpha)} ,P_\nu^{(\alpha)} \rangle^{\alpha}=0$ if $\mu\neq\nu$.

In fact explicit formulae for $\langle P_\mu^{(\alpha)}, P_\mu^{(\alpha)}\rangle^{\alpha}$ are know, but rather complicated, so I won't list them here.

Question: What happens when I vary the parameter $\alpha$ of the inner product, but not for the Jack polynomials? Does anyone know how to evaluate $\langle P_\mu^{(\alpha)} ,P_\nu^{(\alpha)} \rangle^{\beta}$, $\beta\in\mathbb{C}$? I am most interested in the case $\beta=-2\alpha$.

  • 1
    $\begingroup$ Have you some data? $\endgroup$ Commented Jun 26, 2013 at 16:55
  • $\begingroup$ What do you mean by data? Examples? $\endgroup$
    – Gonneman
    Commented Jul 31, 2013 at 8:06
  • $\begingroup$ Yes, examples, do you get nice numbers? Why are you interested in this in particular? $\endgroup$ Commented Mar 3, 2016 at 1:39

1 Answer 1


You ask what happens when you vary the parameter $\alpha$ of the inner product, but not for the Jack polynomials. I cannot say much in the general case, but one instance of interest for me was when the Jack polynomials under consideration are the Schur polynomials ($\alpha=1$). Then the inner product is preserved under duality of partitions $\lambda \mapsto \lambda^*$: $$ \langle J^{(1)}_\lambda,J^{(1)}_\mu \rangle^\beta = \langle J^{(1)}_{\lambda^*},J^{(1)}_{\mu^*} \rangle^\beta. $$

See Inverse Participation Ratios in the XX spin chain for more details.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.