---
title: Bohr Inequality & Lacunary Series
url: https://www.emergentmind.com/topics/bohr-inequality-with-lacunary-series
type: topic
---

# Bohr Inequality & Lacunary Series

to=python code
tool call error: tool must be called with a valid recipient and content matching the tool schema

to=python code
tool call error: tool must be called with a valid recipient and content matching the tool schema

to=shell  彩神争霸高 code
python - <<'PY'
import requests
url='http://export.arxiv.org/api/query?search_query=all:Bohr%20inequality%20lacunary%20series&start=0&max_results=5'
try:
    r=requests.get(url, timeout=20)
    print(r.status_code)
    print(r.text[:500])
except Exception as e:
    print("ERR",e)
PY

to=shell ్యూట code
python - <<'PY'
import requests
url='https://export.arxiv.org/api/query?search_query=all:Bohr%20inequality%20lacunary%20series&start=0&max_results=5'
try:
    r=requests.get(url, timeout=20)
    print(r.status_code)
    print(r.text[:800])
except Exception as e:
    print("ERR",e)
PY

to=shell  天天爱彩票中奖 code
python - <<'PY'
import requests, urllib.parse
q=urllib.parse.quote('Bohr inequality lacunary series')
url='https://arxiv.org/search/?query=%s&searchtype=all&abstracts=show&order=-announced_date_first&size=5' % q
try:
    r=requests.get(url, timeout=20, headers={'User-Agent':'Mozilla/5.0'})
    print(r.status_code)
    print(r.text[:500])
except Exception as e:
    print("ERR",e)
PY

to=shell 下载彩神争霸 code
python - <<'PY'
import requests, bs4, urllib.parse, re
q=urllib.parse.quote('Bohr inequality lacunary series')
url='https://arxiv.org/search/?query=%s&searchtype=all&abstracts=show&order=-announced_date_first&size=10' % q
html=requests.get(url,headers={'User-Agent':'Mozilla/5.0'},timeout=20).text
print('len',len(html))
soup=bs4.BeautifulSoup(html,'html.parser')
results=soup.select('li.arxiv-result')
for i,res in enumerate(results[:10],1):
    title=res.select_one('p.title')
    pid=res.select_one('p.list-title a')
    if title and pid:
        print(i, pid.text.strip(), '-', ' '.join(title.text.split())[:120])
PY

Bohr inequality with lacunary series concerns Bohr-type majorant estimates for analytic or holomorphic functions whose nonzero coefficients occur only on a sparse set of degrees, most commonly an arithmetic progression such as \(\{kp+m:k\ge 0\}\) or \(\{qs+m:s\ge 1\}\). In the classical one-variable form, if \(f(z)=\sum_{n\ge 0}a_n z^n\) is analytic on \(\mathbb D\) and \(\|f\|_\infty\le 1\), then \(\sum_{n\ge 0}|a_n|r^n\le 1\) for \(0\le r\le 1/3\), and \(1/3\) is sharp. The lacunary variant asks how this radius changes when many degrees are excluded. Recent work shows that arithmetic-progression lacunarity frequently enlarges the admissible radius, and that the phenomenon persists in finite-dimensional Banach sequence spaces, for vector-valued holomorphic mappings, and for refined Bohr–Rogosinski functionals with square-sum or energy terms [1708.05578] [2404.18623] [2409.16610] [2509.03532].

## 1. Classical baseline and the meaning of lacunarity

The classical Bohr inequality is the reference point for all subsequent developments. For \(f(z)=\sum_{n=0}^\infty a_n z^n\in H(\mathbb D,\mathbb D)\), the majorant series satisfies
\[
\sum_{n=0}^{\infty}|a_n|r^n\le 1 \qquad (0\le r\le 1/3),
\]
and the constant \(1/3\) is sharp. In the lacunary setting, one restricts the support of the expansion to a sparse family of degrees. In the recent Banach-space literature, the dominant model is arithmetic-progression lacunarity,
\[
\Lambda_{p,m}=\{sp+m:s\in\mathbb N_0\},
\]
or equivalently \(\Lambda=\{qs+m:s\ge 1\}\), with \(p,q\in\mathbb N\) and \(m\in\mathbb N_0\). This includes odd, even, and fixed congruence-class degrees as special cases [2404.18623] [2509.03532].

This notion is distinct from general Hadamard-gap lacunarity. Several of the cited papers explicitly emphasize that the results are proved for arithmetic-progression support rather than for arbitrary sparse sets. A recurring theme is that removing intermediate degrees reduces the cumulative size of the majorant, so the Bohr radius can exceed the classical value \(1/3\). In one variable this enlargement can be dramatic: for bounded analytic odd functions, corresponding to \((p,m)=(2,1)\), the sharp Bohr radius becomes \(0.78991\ldots\) [1708.05578].

## 2. Banach-space and vector-valued framework

The modern formulations are expressed in terms of Fréchet expansions on Banach balls. A standard setting is the finite-dimensional complex Banach sequence space \(E=\ell_t^n\), \(1\le t\le\infty\), with norm
\[
\|z\|_t=\Big(\sum_{j=1}^n|z_j|^t\Big)^{1/t}\quad (1\le t<\infty), 
\qquad 
\|z\|_\infty=\max_{1\le j\le n}|z_j|,
\]
and open unit ball
\[
B_{\ell_t^n}=\{z\in\mathbb C^n:\|z\|_t<1\}.
\]
A holomorphic mapping \(F\in \mathcal H(B_X,Y)\) has the Fréchet expansion
\[
F(z)=\sum_{s=0}^\infty \frac{D^sF(0)(z^s)}{s!},
\]
where \(D^sF(0)\) is the \(s\)-th Fréchet derivative, viewed as a continuous symmetric \(s\)-linear map [2509.03532] [2409.16610].

For vector-valued Bohr inequalities, a common target is the closed unit polydisk \(\overline{\mathbb D^n}\subset\mathbb C^n\) endowed with the sup-norm
\[
\|w\|_\infty=\max_{1\le j\le n}|w_j|.
\]
Lacunarity is then imposed degreewise on the homogeneous pieces in the Fréchet expansion. In this formulation, a vector-valued lacunary series has the form
\[
f(z)=a_0+\sum_{s=0}^\infty A_{sp+m}(z)
\]
or, equivalently,
\[
f(z)=\frac{1}{m!}D^m f(0)(z^m)+\sum_{s=1}^\infty \frac{1}{(sp+m)!}D^{sp+m}f(0)(z^{sp+m}).
\]
The geometric input is usually a Schwarz mapping \(v:B_X\to B_Y\) with a zero of order \(k\) at the origin, for which the Banach-space Schwarz lemma yields
\[
\|v(z)\|_Y\le \|z\|_X^k.
\]
That estimate is fundamental in mixed Bohr and Bohr–Rogosinski inequalities because it converts the order of vanishing of \(v\) into an effective reduction of the radial parameter [2509.03532].

## 3. Sharp radii for arithmetic-progression lacunary series

The best-known scalar one-variable theorem in this direction concerns bounded analytic functions of the form
\[
f(z)=\sum_{k=0}^{\infty}a_{kp+m}z^{kp+m}, \qquad 0<m<p.
\]
The sharp radius \(r_{p,m}\) is the largest positive solution of
\[
-6r^{p-m}+r^{2(p-m)}+8r^{2p}+1=0,
\]
and
\[
\sum_{k=0}^{\infty}|a_{kp+m}|r^{kp+m}\le 1 \qquad (0\le r\le r_{p,m}).
\]
For \(m=0\), the sharp radius is \(r_{p,0}=1/3\); if in addition \(a_0=0\), the radius increases to \(1/\sqrt 2\) [1708.05578].

In higher-dimensional sequence spaces, exact sharp radii also arise for arithmetic lacunarity. For scalar- and vector-valued holomorphic mappings on \(B_{\ell_t^n}\), under the structural hypotheses used in the paper, the sharp radius is the unique root in \((0,1)\) of
\[
r^{p+m}+r^{2p}-1=0.
\]
The same equation governs both the scalar-target and vector-valued lacunary inequalities in that setting, and the radii are dimension-free in the sense that they do not explicitly depend on \(n\) [2404.18623].

A later refined Banach-space formulation replaces the basic majorant by a sum with squared-coefficient corrections. In that setting the sharp lacunary radius \(r_{p,m}^*\) is the maximal positive root of
\[
5r^{2p+m}-2r^{p+m}+r^m+4r^{2p}-4r^p=0.
\]
For \(m=0\), one has \(r_{p,0}^*=1/\sqrt 3\) for every \(p\) [2409.16610].

A representative comparison of sharp radii is given below.

| Setting | Sharp radius or defining equation | Source |
|---|---|---|
| Analytic \(f(z)=\sum a_{kp+m}z^{kp+m}\), \(0<m<p\) | \(-6r^{p-m}+r^{2(p-m)}+8r^{2p}+1=0\) | [1708.05578] |
| Lacunary scalar/vector on \(B_{\ell_t^n}\) | \(r^{p+m}+r^{2p}-1=0\) | [2404.18623] |
| Refined lacunary Bohr on \(B_X\) | \(5r^{2p+m}-2r^{p+m}+r^m+4r^{2p}-4r^p=0\) | [2409.16610] |
| Mixed vector-valued corollary | \(r\le 1/5\) | [2509.03532] |

These formulas make explicit the quantitative effect of sparsity. As the gap parameter increases, the radius equations involve higher powers of \(r\), and the admissible radius typically grows. This suggests that arithmetic lacunarity acts as a structural damping mechanism on the majorant series.

## 4. Vector-valued, mixed, and refined Bohr inequalities

The 2025 vector-valued treatment on \(B_{\ell_t^n}\) studies holomorphic mappings
\[
f:B_{\ell_t^n}\to \overline{\mathbb D^n}, \qquad 
f(z)=\sum_{s=0}^{\infty}\frac{D^s f(0)(z^s)}{s!},
\]
with \(a=f(0)\) and \(b=\|a\|_\infty\). Its principal mixed theorem combines three ingredients: a point evaluation term \(\|f(v_1(z))\|_\infty^p\), a lacunary majorant over degrees \(qs+m\), and a shift term \(\|f(v_2(z))-f(0)\|_\infty\), where \(v_1\) and \(v_2\) are Schwarz mappings of prescribed orders. The resulting radius is sharp and is characterized as the minimal root of an explicit algebraic equation in \(r\) involving the parameters \(p,q,m,m_1,m_2,\mu,\nu\). A corollary gives
\[
\sum_{s=0}^{\infty}\frac{\|D^s f(0)(z^s)\|_\infty}{s!}
+\|f(z)-f(0)\|_\infty \le 1
\qquad \text{for } r\le \frac15,
\]
and \(1/5\) is sharp [2509.03532].

Refined versions add square-sum corrections. One such functional is
\[
\mathcal N_f^1(r)=\sum_{s=1}^{\infty}\frac{\|D^s f(0)(z^s)\|_\infty}{s!}
+\left(\frac{1}{1+\|f(0)\|_\infty}+\frac{r}{1-r}\right)
\sum_{s=1}^{\infty}\left(\frac{\|D^s f(0)(z^s)\|_\infty}{s!}\right)^2.
\]
Together with a Schwarz-shift term, this yields a sharp refined Bohr inequality up to a radius \(R_2(p)\) determined by an explicit root equation [2509.03532].

A related Banach-space theory for scalar-valued mappings \(f:B_X\to \mathbb D\) and vector-valued mappings \(f:B_X\to Y\) introduces refined sums such as
\[
|f(0)|+\sum_{s=1}^{\infty}|P_s(z)|
+\frac{1+|f(0)|}{1-r}\sum_{s=1}^{\infty}|P_s(z)|^2\le 1
\qquad (r\le 1/3),
\]
with sharp constant \(1/3\), and analogous functional-type and norm-type lacunary versions governed by the radius \(r_{p,m}^*\) above [2409.16610]. Earlier multidimensional work on \(\mathbb D^n\) and on balanced Banach domains established sharp refined norm-type inequalities under restricted coordinate hypotheses, including the sharp constant \(3/5\) for general lacunary expansions without a constant term [2303.08855].

## 5. Bohr–Rogosinski, alternating variants, and energy functionals

The Bohr–Rogosinski extension controls a tail majorant together with a point evaluation. In the vector-valued Banach-sequence-space setting, the basic Rogosinski functional \(\mathcal M_f^N(r)\) starts at degree \(N\) and satisfies the sharp estimate
\[
\mathcal M_f^N(r)\le \frac{(1-\|f(0)\|_\infty^2)\,r^N}{1-r}.
\]
This leads to inequalities of the form
\[
\|f(v_1(z))\|_\infty^p+\mathcal M_f^N(r)\le 1,
\]
with a sharp radius defined as the unique root of an explicit equation in \(r\) depending on \(p\), \(N\), and the order of vanishing of the Schwarz mapping \(v_1\) [2509.03532]. An analogous scalar Banach-space theorem appears in the refined 2024 paper, again with a sharp root equation and the classical special cases \(r\le 1/3\) for \((p,N)=(1,1)\) and \(r\le 1/2\) for \((p,N)=(2,1)\) [2409.16610].

Another family of refinements augments the majorant by an energy term. In the vector-valued setting,
\[
S_z=\sum_{s=1}^{\infty} s\left(\frac{\|D^s f(0)(z^s)\|_\infty}{s!}\right)^2
\]
is combined with a polynomial \(W_N(S_z)\). Under an explicit coefficient condition on \(W_N\), one obtains
\[
\mathcal C_f(r)=\|f(0)\|_\infty^p+\mathcal N_f^1(r)+W_N(S_z)\le 1
\]
for \(r\le p/(2+p)\), and the radius is sharp. In particular, for \(p=N=1\), \(d_1=8/9\), and \(d_j=0\) for \(j\ge 2\),
\[
\|f(0)\|_\infty+\mathcal N_f^1(r)+\frac89\,S_z\le 1
\qquad \text{for } r\le \frac13,
\]
with sharp radius \(1/3\) [2509.03532].

Alternating lacunary variants form a parallel line of development. For \(p\) odd and \(1<m<p\), vector-valued alternating lacunary inequalities on \(B_{\ell_t^n}\) hold with the same sharp radius determined by
\[
r^{p+m}+r^{2p}-1=0,
\]
while a mixed alternating majorant yields the sharp radius defined by
\[
r^{2p+m}+2r^{2p}-1=0.
\]
Weighted one-variable alternating theories for arithmetic-progression subseries produce further sharp root formulas such as \(r^{2p}+r^{p+m}-1=0\) [2404.18623] [2106.11158].

## 6. Extremals, dimension effects, limitations, and open problems

Sharpness is established throughout by explicit Möbius-type extremals. In one variable and in arithmetic-progression classes, the standard extremal is
\[
f_*(z)=z^m\frac{z^p-a}{1-a z^p},
\]
or equivalent Blaschke-type variants. These attain equality in the sharp radius calculations for scalar lacunary Bohr inequalities [1708.05578]. In Banach spaces, extremals are built from norm-attaining functionals \(T_w\in X^*\) and the same Möbius mechanism,
\[
f_a(z)=\frac{a+T_w(z)}{1+a\,T_w(z)},
\qquad
f_{a,p,m}(z)=(T_w(z))^m\frac{(T_w(z))^p-a}{1-a(T_w(z))^p},
\]
which reduce the problem to a one-dimensional slice [2409.16610]. For vector-valued maps into \(\mathbb C^n\), an extremal family is
\[
F(z)=\left(\frac{b+z_1}{1+b z_1},\,0,\dots,0\right),
\]
and evaluation at \(z=(r,0,\dots,0)\) shows failure beyond the asserted radius as \(b\to 1^{-}\) [2509.03532].

A notable structural feature of many recent results is their lack of explicit dimension dependence. In several theorems on \(B_{\ell_t^n}\), the radii depend on lacunarity parameters and on the order of vanishing of auxiliary Schwarz mappings, but not directly on \(n\). This is attributed to estimates along a single coordinate or along the scalar radial parameter \(r=\|z\|_t\) rather than on full coefficient counting in dimension \(n\) [2404.18623] [2509.03532]. By contrast, unrestricted multidimensional Bohr radii \(K_n\) for the polydisk satisfy the known asymptotic behavior
\[
K_n\sim \sqrt{\frac{\log n}{n}},
\]
which shows that the lacunary and restricted-direction setting is qualitatively different [2509.03532].

The main limitations are equally clear. The strongest sharp results are proved for finite-dimensional sequence spaces \(E=\ell_t^n\), for targets such as \(\mathbb C^n\) with the sup-norm, and for arithmetic-progression lacunarity. Broader sparse sets of degrees, infinite-dimensional domains, and general Banach targets remain largely open. Existing papers explicitly identify extensions to infinite-dimensional sequence spaces, to Banach-valued targets beyond \(\mathbb C^n\), and to more general lacunary patterns such as Hadamard gaps as nontrivial problems that would likely require additional geometric hypotheses or stronger hypercontractive inequalities [2404.18623] [2509.03532].

Source: https://www.emergentmind.com/topics/bohr-inequality-with-lacunary-series