scientific article; zbMATH DE number 2063246

From MaRDI portal
Publication:4460865





In this succinct and interesting note the author continues earlier work of \textit{M. Kaneko} and \textit{D. Zagier} [AMS/IP Stud. Adv. Math. 7, 97--126 (1998; Zbl 0955.11018)] and \textit{M. Kaneko} and \textit{N. Todaka} [CRM Proc. Lect. Notes 30, 79--83 (2001; Zbl 1037.11030)] on differential endomorphisms of the spaces \(M_{2k}(\Gamma_{(1)})\) (entire modular forms of weight \(2k\), \(k\in\mathbb{Z}\), on the full modular group). From the author's introduction: ``In this simple note we study a similar construction for the spaces of modular forms \(M_k(\Gamma_0(4),\chi_k)\) where \(k\in {1\over 2}\mathbb{N}\) and \[ \chi_k:= \begin{cases} \chi_{\text{triv}}\quad &\text{if }k\in 2\mathbb{Z}\text{ or }k\in{1\over 2}+ \mathbb{Z},\\ \chi_{-4}\quad &\text{if }k\in 1+ 2\mathbb{Z}.\end{cases} \] We construct, uniformly in \(k\), endomorphisms on \(M_k(\Gamma_0(4),\chi_k)\). We characterize some of their intrinsic properties which may be of interest to those working on the combinatorial aspects of \(q\)-series and hypergeometric functions. ``We shall identify a modular form \(f(z)\in M_k(\Gamma_0(4),\chi_k)\) with its Fourier expansion \[ f(z)= \sum^\infty_{n=0} a(n)q^n\qquad (q:= e^{2\pi iz}\text{ throughout}). \] For convenience, let \(d\) denote the differential operator defined by \[ d\Biggl(\sum^\infty_{n=0} a(n)q^n\Biggr)= \sum^\infty_{n=0} na(n)q^n= {1\over 2\pi i}\cdot{d\over dz} \Biggl(\sum^\infty_{n=0} a(n) e^{2\pi inz}\Biggr). \] If \(k\in{1\over 2}\mathbb{N}\), then let \(C_k\) denote the map \[ C_k: M_k(\Gamma_0(4), \chi_k)\to \mathbb{C}q \] given by \[ C_k(f):= {(k+ 1/2)(f\cdot d(\theta))-{1\over 2} d(f\theta)\over F\theta}, \] where \[ \theta(z)= \sum^\infty_{n=-\infty} q^{n^2}= 1+ 2q+ 2q^4+ 2q^9+\cdots, \] and \[ F(z)= \sum^\infty_{n=0} \sigma_1(2n+ 1)q^{2n+1}. \] Theorem 1. If \(k\in{1\over 2} \mathbb{N}\), then the map \(C_k\), \[ C_k: M_k(\Gamma_0(4), \chi_k)\to M_k(\Gamma_0(4), \chi_k) \] is a \(\mathbb{C}\)-linear endomorphism. Moreover, the kernel of \(C_k\) is \(\ker(C_k)= \mathbb{C}\theta^{2k}\).\dots ``Definition. A modular form \(f\in M_k(\Gamma_0(4), \chi_k)\) is a normalized eigenform if \hskip17mm (1) The first non-zero coefficient in the \(q\)-series expansion is 1. \hskip17mm (2) There is a complex number \(\lambda_f\) for which \(C_k= \lambda_ff\). Theorem 2. If \(k\in{1\over 2}\mathbb{N}\), then there are exactly \(e:=\dim(M_k(\Gamma_0(4), \chi_k)- 1\) normalized eigenforms, say \(f_{1,k},f_{2,k},\dots, f_{e,k}\), of the endomorphism \(C_k\). If \(e\geq 1\), then for all \(1\leq m\leq e\) we have \hskip17mm (i) \(f_{m,k}:= (1-{16F\over \theta^4})^m\cdot\theta^{2k}= {_1F_0}(^{-m}\mid 16F\theta^4)\cdot \theta^{2k}\), \hskip17mm (ii) \(C_k(f_{m,k})= 8m f_{m,k}\). It is natural to ask which cusp forms are in the image of \(C_k\). We prove: Theorem 3. If \(k\in{1\over 2}\mathbb{N}\), then we have \(S_k(\Gamma_0(4),\chi_k)\subseteq \text{Im}(C_k)\). For the entire collection see [Zbl 0980.00026].











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4460865)