Skip to main content
I edited the answer to make it actually resolve the question
Source Link
Milo Moses
  • 2.9k
  • 12
  • 36

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, theThe generalized Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$\begin{equation} \hat{E}_{k,\chi,\psi}(z):=\sum_{n=1}^{\infty}\left(\sum_{d|n}\psi(d)\chi(n/d)d^{k-1}\right)q^n\tag{1} \end{equation}

is anIs a eigenform ofover the space $M_k(N,\chi)$$M_k(RL,\chi\psi)$ for characters $\psi$ and $\chi$ with conductors $L$ and $R$ respectively, whenever $L>1$ and $\psi(-1)\chi(-1)=(-1)^k$. ThusMore generally, lettingfor any $\chi=\chi_2$$t>0$ the function $\hat{E}_{k,\chi,\psi}\left(z^t\right)$ is a modular form over $M_k(RLt,\chi\psi)$ according to this online textbook, Theorem 5.8.

We now set $\psi=1$ to be the trivial character and $\chi_=\chi_2$ to be the unique Dirichlet character modulo $2$. This means that $L=2>1$ and $\psi(-1)\chi(-1)=1$ so the conditions of (i.e1) are satisfied if $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$$k$ is even, making $\hat{E}_{k,\chi_2,1}(z)$ a modular (eigen) form. Now, if $k'$ is any odd integer we get thatcan let $k=1-k'$ be an even integer, so

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}\begin{equation} \hat{E}_{1-k',\chi_2,1}(z)=\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(n/d)d^{-k'}\right)q^n \end{equation}

Thus, we can differentiate our functionsis a modular form of degree $f_k(z)$ repeatedly$1-k'$ over $k$ times to get$\Gamma_1(2)$. Since $\sum_{d|n}\chi_2(n/d)d^{-k'}=\frac{\sigma^{(o)}_{k}(n)}{n^k}$, this means that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}\begin{equation} \hat{E}_{1-k,\chi_2,1}(z)=\sum_{n=1}^{\infty}\frac{\sigma^{(o)}_{k}(n)}{n^k}q^n\tag{2} \end{equation}

and thusRelating this to $f_{k}(z)$ is now a straightforward task. Namely, we see that

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$\begin{align*} f_k(2z+1)&=2i\sum_{n=1}^{\infty}\frac{(-1)^{n+1}\sigma_k^{(o)}(n)}{n^k}\exp\left(\frac{2\pi in(2z+1)}{2}\right)\\ &=-2i\sum_{n=1}^{\infty}\frac{\sigma_k^{(o)}(n)}{n^k}\exp\left(2\pi inz\right)\\ &=-2i\hat{E}_{1-k,\chi_2,1}(z) \end{align*}

In a way this explains the connection we are seeingThus, but in another it makes it even more mysterious since there is no reason that the iterated integral of antransformation $f_k(2z+1)i/2$ turns $f_k(z)$ into a eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is Working out that $f_k(z)$$f_k(2z+1)i/2$ is an eigenform just from the function generated byformulae given in the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$problem statement seems hard, and so perhaps for any choicebut also I am not aware of Dirichlet character $\chi$any general formula for Eisenstein series which yields the $q$-series generated by $\sum_{d|n}\chi\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties,formulae given in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$problem statement.

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$I find this stuff absolutely fascinating, and I'll be looking more into this and undoubtedly coming up with lots more fun stuff, so if anyone actually ends up reading this and cares about I'm writting then I will add more answers/edit this answer as time goes on to make the solution more complete. Thank you for your time.

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

The generalized Eisenstein series

\begin{equation} \hat{E}_{k,\chi,\psi}(z):=\sum_{n=1}^{\infty}\left(\sum_{d|n}\psi(d)\chi(n/d)d^{k-1}\right)q^n\tag{1} \end{equation}

Is a eigenform over the space $M_k(RL,\chi\psi)$ for characters $\psi$ and $\chi$ with conductors $L$ and $R$ respectively, whenever $L>1$ and $\psi(-1)\chi(-1)=(-1)^k$. More generally, for any $t>0$ the function $\hat{E}_{k,\chi,\psi}\left(z^t\right)$ is a modular form over $M_k(RLt,\chi\psi)$ according to this online textbook, Theorem 5.8.

We now set $\psi=1$ to be the trivial character and $\chi_=\chi_2$ to be the unique character modulo $2$. This means that $L=2>1$ and $\psi(-1)\chi(-1)=1$ so the conditions of (1) are satisfied if $k$ is even, making $\hat{E}_{k,\chi_2,1}(z)$ a modular (eigen) form. Now, if $k'$ is any odd integer we can let $k=1-k'$ be an even integer, so

\begin{equation} \hat{E}_{1-k',\chi_2,1}(z)=\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(n/d)d^{-k'}\right)q^n \end{equation}

is a modular form of degree $1-k'$ over $\Gamma_1(2)$. Since $\sum_{d|n}\chi_2(n/d)d^{-k'}=\frac{\sigma^{(o)}_{k}(n)}{n^k}$, this means that

\begin{equation} \hat{E}_{1-k,\chi_2,1}(z)=\sum_{n=1}^{\infty}\frac{\sigma^{(o)}_{k}(n)}{n^k}q^n\tag{2} \end{equation}

Relating this to $f_{k}(z)$ is now a straightforward task. Namely, we see that

\begin{align*} f_k(2z+1)&=2i\sum_{n=1}^{\infty}\frac{(-1)^{n+1}\sigma_k^{(o)}(n)}{n^k}\exp\left(\frac{2\pi in(2z+1)}{2}\right)\\ &=-2i\sum_{n=1}^{\infty}\frac{\sigma_k^{(o)}(n)}{n^k}\exp\left(2\pi inz\right)\\ &=-2i\hat{E}_{1-k,\chi_2,1}(z) \end{align*}

Thus, the transformation $f_k(2z+1)i/2$ turns $f_k(z)$ into a eigenform. Working out that $f_k(2z+1)i/2$ is an eigenform just from the formulae given in the problem statement seems hard, but also I am not aware of any general formula for Eisenstein series which yields the formulae given in the problem statement.

I find this stuff absolutely fascinating, and I'll be looking more into this and undoubtedly coming up with lots more fun stuff, so if anyone actually ends up reading this and cares about I'm writting then I will add more answers/edit this answer as time goes on to make the solution more complete. Thank you for your time.

deleted 2 characters in body
Source Link
Milo Moses
  • 2.9k
  • 12
  • 36

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$$\sum_{d|n}\chi\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

Added a second way to look at what is going on, hinting at a more general situation.
Source Link
Milo Moses
  • 2.9k
  • 12
  • 36

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

I think I am starting to understand what is going on on. For any Dirichlet character $\chi$, the Eisenstein series

$$\hat{E}_{k,\chi}(z)=-\frac{B_{k,\chi}}{2k}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi(d)d^{k-1}\right)q^n$$

is an eigenform of the space $M_k(N,\chi)$. Thus, letting $\chi=\chi_2$ be the unique Dirichlet character modulo $2$ (i.e $\chi_2(n)=n\,\,\mathrm{mod}\,\,2$) we get that

\begin{align*} \hat{E}_{k+1,\chi_2}(z)&=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\left(\sum_{d|n}\chi_2(d)d^{k}\right)q^n\\ &=-\frac{B_{k+1,\chi_2}}{2(k+1)}+\sum_{n=1}^{\infty}\sigma^{(o)}_{k}(n)q^n \end{align*}

Thus, we can differentiate our functions $f_k(z)$ repeatedly $k$ times to get that

\begin{align*} f^{(k)}_k(z)&=\frac{d^k}{dz^{k}}\sum_{n=1}^{\infty}\frac{(-1)^{n-1}\sigma_k^{(o)}(n)}{n^k}e^{z\pi i n}\\ &=(i\pi)^k\sum_{n=1}^{\infty}(-1)^{n-1}\sigma_k^{(o)}(n)e^{z\pi i n}\\ &=-(i\pi)^k\sum_{n=1}^{\infty}\sigma_k^{(o)}(n)e^{(z-1)\pi i n}\\ \end{align*}

and thus

$$\frac{-f^{(k)}_k(z+1)}{(i\pi)^k}=\hat{E}_{k+1,\chi_2}(z)+\frac{B_{k+1,\chi_2}}{2(k+1)}$$

In a way this explains the connection we are seeing, but in another it makes it even more mysterious since there is no reason that the iterated integral of an eigenform should retain its nice properties.

Another (possibly more enlightening) way to think about what is going on is that $f_k(z)$ is the function generated by the $q$-series $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$, and so perhaps for any choice of Dirichlet character $\chi$ the $q$-series generated by $\sum_{d|n}\chi_2\left(\frac{n}{d}\right)d^{-k}$ will have modular form-like properties, in the sense that

$$z^{\frac{1-k}{2}}g(z)+z^{\frac{k-1}{2}}g\left(\frac{1}{z}\right)=\mathrm{poly}(z)$$

where $\mathrm{poly}(z)$ is a polynomial of degree at most $\frac{k+1}{2}$.

Source Link
Milo Moses
  • 2.9k
  • 12
  • 36
Loading