A general formula for Hecke-type false theta functions
Eric T. Mortenson
TL;DR
The paper addresses expressing Hecke-type double-sums $f_{a,b,c}(x,y;q)$ in the regime $D=b^2-ac<0$ as combinations of theta and false theta functions, extending Matsusaka's decompositions. It develops a general decomposition (Theorem mainTheorem) by leveraging elliptic transformation properties and the Jacobi triple product, and then validates the framework with explicit Habiro-type examples such as $H_p^{(2)}(q)$, including concrete cases like $H_{1}^{(2)}(q)$ and $H_{2}^{(2)}(q)$. The results provide a unified approach that mirrors the Appell-theta decompositions known for $D>0$ while yielding convergent partial-theta structures for $D<0$. This advances understanding of the modular-like structure and radial limits of Habiro-type series by linking double-sums to theta and false theta building blocks.
Abstract
In recent work where Matsusaka generalizes the relationship between Habiro-type series and false theta functions after Hikami, five families of Hecke-type double-sums of the form \begin{equation*} \left( \sum_{r,s\ge 0 }-\sum_{r,s<0}\right)(-1)^{r+s}x^ry^sq^{a\binom{r}{2}+brs+c\binom{s}{2}}, \end{equation*} where $b^2-ac<0$, are decomposed into sums of products of theta functions and false theta functions. Here we obtain a general formula for such double-sums in terms of theta functions and false theta functions, which subsumes the decompositions of Matsusaka. Our general formula is similar in structure to the case $b^2-ac>0$, where Mortenson and Zwegers obtain a decomposition in terms of Appell functions and theta functions.
