Difference between revisions of "Nemtsov function"

From TORI
Jump to navigation Jump to search
(import from Sandbox)
Line 1: Line 1:
 
{{top}}
 
{{top}}
 
<div class="thumb tright" style="float:right; margin:-76px 0px 0px 8px; line-height:2px; background-color:#fff">
 
<div class="thumb tright" style="float:right; margin:-76px 0px 0px 8px; line-height:2px; background-color:#fff">
{{pic|Nemplot.jpg|120px}}<small><center>\(y=\mathrm{nem}_q(x)\)<!--<br> for various \(q\)!--></center></small>
+
{{pic|Nemplot.jpg|120px}}<small><center>\(y=\mathrm{nem}_q(x)\)</center></small>
 
</div>
 
</div>
   
 
<div class="thumb tright" style="float:right; margin:0px 0px 4px 8px; width:408px">
 
<div class="thumb tright" style="float:right; margin:0px 0px 4px 8px; width:408px">
 
{{pic|Nem0map.jpg|200px}} {{pic|ArqNem0map.jpg|200px}}
 
{{pic|Nem0map.jpg|200px}} {{pic|ArqNem0map.jpg|200px}}
  +
{{pic|Nem2map.jpg|200px}} {{pic|ArqNem2map.jpg|200px}}
 
{{pic|Nem2map.jpg|200px}} {{pic|ArqNem2map.jpg|200px}}<small><center>Complex maps: \(u\!+\!\mathrm i v=f(x\!+\!\mathrm i y)\) for \(f\!=\!\mathrm{nem}_0\), \(f\!=\!\mathrm{ArqNem}_0 ~~\) (top) and for \(f\!=\!\mathrm{nem}_2\), \(f\!=\!\mathrm{ArqNem}_2\) (Bottom)</center></small>
+
<small><center>Complex maps: \(u+\mathrm i v=f(x+\mathrm i y)\) for \(f=\mathrm{nem}_0\), \(f=\mathrm{ArqNem}_0\) (top) and for \(f=\mathrm{nem}_2\), \(f=\mathrm{ArqNem}_2\) (bottom)</center></small>
 
</div>
 
</div>
   
Line 14: Line 14:
 
https://www.morebooks.de/shop-ui/shop/product/978-620-2-67286-3 <br>
 
https://www.morebooks.de/shop-ui/shop/product/978-620-2-67286-3 <br>
 
https://mizugadro.mydns.jp/BOOK/468.pdf <br>
 
https://mizugadro.mydns.jp/BOOK/468.pdf <br>
Dmitrii Kouznetsov. [[Superfunctions]]. [[Lambert Academic Publishing]], 2020.
+
Dmitrii Kouznetsov. ''Superfunctions''. [[Lambert Academic Publishing]], 2020.
</ref>, 2020.<br>
+
</ref>.
The description is loaded also as the [[Mizugadro Preprint]] <ref name="prep1015">
+
The description is also available as the [[Mizugadro Preprint]] <ref>
https://mizugadro.mydns.jp/PAPERS/2016nemtsov.pdf
+
https://mizugadro.mydns.jp/PAPERS/2016nemtsov.pdf <br>
Dmitrii Kouznetsov.
+
Dmitrii Kouznetsov. ''Nemtsov function and its iterates''. [[Mizugadro Preprint]], 2016.
Nemtsov function and its iterates.
 
[[Mizugadro Preprint]], 2016.
 
 
</ref>.
 
</ref>.
   
The [[Nemtsov function]] \(y=\mathrm{nem}_q(x)= x+x^3+q x^4\)
+
The [[Nemtsov function]] \(y=\mathrm{nem}_q(x)= x+x^3+q\,x^4\) is shown in the figure at right for various \(q\ge 0\).
is shown in figure at right versus \(x\) for various \(q\!\ge\!0\).
 
   
[[Complex map]]s of \( \mathrm{nem}_q \) are shown in the left-hand column of the figure below.
+
[[Complex map]]s of \(\mathrm{nem}_q\) are shown in the left column of the figure below.
  +
The right column shows similar maps for the inverse function \(\mathrm{ArqNem}_q=\mathrm{nem}_q^{-1}\).
   
  +
This article describes the Nemtsov function and its related functions:
The right-hand column shows similar maps for the inverse function \( \mathrm{ArqNem}_q = \mathrm{nem}_q^{-1} \).
 
  +
the inverse function [[ArqNem]],
  +
the [[superfunction]] [[SuNem]], and
  +
the [[Abel function]] [[AuNem]].
   
  +
== Motivation ==
This article describes the Nemtsov function nem and related functions:<br>
 
the inverse function [[ArqNem]],<br>
 
the [[superfunction]] [[SuNem]] and<br>
 
the [[abelfunction]] [[AuNem]].
 
   
  +
The [[Nemtsov function]] serves as an example of a real-holomorphic [[transfer function]] with a real [[fixed point]], where the usual [[regular iteration]] method for constructing a [[superfunction]] cannot be applied directly.
==Motivation and naming==
 
  +
The editor did not find any simpler example of this type other than this specific fourth-order polynomial.
   
The [[Nemtsov function]] serves as a simple example of a real-holomorphic [[transfer function]] with a real [[fixed point]], where the [[regular iteration]] method for constructing a [[superfunction]] cannot be applied "as is".<br> The editor did not find any simpler example of this kind, other than the special polynomial of 4th order.
+
The function was introduced as an attempt to construct an “exotic’’ [[transfer function]] for which a growing real-holomorphic [[superfunction]] could not be produced by the usual methods appearing in the «[[Table of superfunctions]]».
  +
This attempt failed — the [[superfunction]] for the Nemtsov function *can* be constructed, and is described below.
   
  +
The expansion of the Nemtsov function at its fixed point begins with a linear term whose coefficient is unity. This prevents the use of standard [[regular iteration]] <ref>
The [[Nemtsov function]] is an attempt to propose an "exotic" [[transfer function]] for which the [[superfunction]] cannot be constructed in the same way as for other [[superfunctions]] from the «[[Table of superfunctions]]».
 
  +
http://www.ams.org/journals/mcom/2010-79-271/S0025-5718-10-02342-2/home.html <br>
  +
https://mizugadro.mydns.jp/PAPERS/2010sqrt2.pdf <br>
  +
D. Kouznetsov, H. Trappmann. ''Portrait of the four regular super-exponentials to base sqrt(2)''. ''Mathematics of Computation'', 2010, v.79, p.1727–1756.
  +
</ref>, which works for the exponential to base \(\sqrt{2}\).
   
  +
The situation is similar to the exponential to base \(\exp(1/\mathrm e)\) <ref name="e1e">
The expansion of the [[Nemtsov function]] at its fixed point begins with
 
the linear term with coefficient unity. This makes impossible the [[Regular iteration]]
 
<ref>
 
http://www.ams.org/journals/mcom/2010-79-271/S0025-5718-10-02342-2/home.html<br>
 
https://mizugadro.mydns.jp/PAPERS/2010sqrt2.pdf
 
D.Kouznetsov, H.Trappmann. Portrait of the four regular super-exponentials to base sqrt(2). [[Mathematics of Computation]], 2010, v.79, p.1727-1756.
 
</ref>. (That works well for the exponential to base \(\sqrt{2}\) as the [[transfer function]]).
 
 
This case is similar to that of the exponential
 
to base \(\exp(1/\mathrm e)\) <ref name="e1e">
 
 
http://www.ams.org/journals/mcom/0000-000-00/S0025-5718-2012-02590-7/S0025-5718-2012-02590-7.pdf <br>
 
http://www.ams.org/journals/mcom/0000-000-00/S0025-5718-2012-02590-7/S0025-5718-2012-02590-7.pdf <br>
https://mizugadro.mydns.jp/PAPERS/2012e1eMcom2590.pdf
+
https://mizugadro.mydns.jp/PAPERS/2012e1eMcom2590.pdf <br>
H.Trappmann, D.Kouznetsov. Computation of the Two Regular Super-Exponentials to base exp(1/e). [[Mathematics of Computation]], v.81 (2012), p. 2207-2227. ISSN 1088-6842(e) ISSN 0025-5718(p)
+
H. Trappmann, D. Kouznetsov. ''Computation of the Two Regular Super-Exponentials to base exp(1/e)''. ''Mathematics of Computation'', 81 (2012), 2207–2227.
</ref>, but has no quadratic term, so, the [[exotic iteration]] <ref name="e1e"/> cannot be applied "as is".
+
</ref>, but the Nemtsov function lacks the quadratic term, so the [[exotic iteration]] of <ref name="e1e"/> cannot be used as is.
   
The case of sin as the [[transfer function]]
+
A closer analogy is the case of the sine function <ref>
<ref>
 
 
http://www.pphmj.com/references/8246.htm <br>
 
http://www.pphmj.com/references/8246.htm <br>
https://mizugadro.mydns.jp/PAPERS/2014susin.pdf<br>
+
https://mizugadro.mydns.jp/PAPERS/2014susin.pdf <br>
Dmitrii Kouznetsov. SUPER SIN.
+
Dmitrii Kouznetsov. ''SUPER SIN''. ''Far East Journal of Mathematical Sciences'' 85(2), 219–238 (2014).
  +
</ref>, but sine is antisymmetric, \(\sin(-z)=-\sin(z)\), which simplifies its analysis.
[[Far East Journal of Mathematical Sciences]] (FJMS)
 
  +
The Nemtsov function for \(q>0\) has no such symmetry, leading initially to doubts whether a superfunction could be constructed.
Volume 85, Issue 2, Pages 219 - 238 (February 2014)
 
</ref> is closer.
 
However, sin is antisymmetric function, \(\sin(-z)=-\sin(z)\).
 
The symmetry simplifies the consideration.<br>
 
The [[Nemtsov function]] \( \mathrm{nem}_q\) at \(q>0\) does not have such a symmetry.
 
<br>
 
This caused doubts, if the superfunction for the [[Nemtsov function]]
 
can be constructed.
 
 
In such a way, the [[Nemtsov function]] appears as an attempt to built-up the growing real-holomorphic function,
 
for which the growing real-holomorphic [[superfunction]] cannot be constructed.
 
   
  +
Eventually, the construction succeeded.
The attempt failed; the [[Superfunction]] for the [[Nemtsov function]] is constructed and described below.
 
   
 
<div class="thumb tright" style="float:right; margin:-4px 0px 4px 8px">
 
<div class="thumb tright" style="float:right; margin:-4px 0px 4px 8px">
{{pic|Boris.jpg|100px}}<small><center>[[Немцов Борис Ефимович|B.Nemtsov]]<ref>
+
{{pic|Boris.jpg|100px}}<small><center>[[Немцов Борис Ефимович|B. Nemtsov]]<ref>http://nemtsov.ru Борис Немцов</ref></center></small></div>
http://nemtsov.ru Борис Немцов </ref></center></small></div>
 
The function name was introduced in 2015 as a personal memorial reference to Boris Nemtsov. The naming is unrelated to the mathematical properties of the function.
 
<br>
 
The need of the special name for this function had been revealed 2015.02.27, in the same day when [[Putin killed Nemtsov]]. (To year 2025, no other [[scientific concept]] of that sad event is found.)
 
Apparently, the total [[corruption]] in Russia <ref name="medvedko">
 
http://kremlin.ru/transcripts/1566 Д.Медведев. Вступительное слово на заседании Совета по противодействию коррупции. 30 сентября 2008 года, 16:25 Москва, Кремль. Д.Медведев: Коррупция в нашей стране приобрела не просто масштабный характер, она стала привычным, обыденным явлением, которое характеризует саму жизнь в нашем обществе...
 
</ref> does not allow the professionals to investigate that case, and that terroristic act caused many publications. This makes the family name "Nemtsov" to be a memorial mark, label on the timeline of Human History.<br>
 
To year 2025, no better notation for this special polynomial is found. So, term «[[Nemtsov function]]» is used here.
 
   
  +
The need for a special name for this function emerged on 2015-02-27, the day when [[Putin killed Nemtsov]].
The function is denoted \( \mathrm{nem}_q \) (lowercase), while “Nem” in “Nemtsov function” is a proper name and not the function symbol.
 
  +
As of 2025, no other scientific concept attached to that event has appeared.
Following tradition to write names of elementary functions with lowercase letters, in formulas it appears as "nem".
 
  +
The total [[corruption]] in Russia <ref name="medvedko">
However, it is capitalized when it is necessary to separate it from the prefix than may indicate the [[inverse function]] ArqNem,
 
  +
http://kremlin.ru/transcripts/1566 <br>
the [[abelfunction]] AuNem or the [[superfunction]] SuNem.
 
  +
D. Medvedev: ''Corruption in our country has acquired not just a large-scale character; it has become a habitual, everyday phenomenon...'' (2008)
  +
</ref> prevents professional investigation of the crime.
  +
Thus, the family name “Nemtsov’’ serves as a historical timestamp.
   
  +
By 2025, no better notation for this polynomial has been proposed.
==Definition and notations==
 
  +
The function symbol is written as lowercase \(\mathrm{nem}\), following the convention for mathematical functions, even though “Nemtsov’’ is a proper name.
  +
Capitalization is used only when distinguishing it from derived functions such as [[ArqNem]], [[AuNem]], and [[SuNem]].
   
  +
== Definition and notations ==
Let \(q\) be a non-negative real parameter.
 
   
  +
Let \(q\ge 0\) be a real parameter.
Then, [[Nemtsov function]] \(\mathrm{nem}\) is defined for complex argument \(z\) as follows:
 
  +
The [[Nemtsov function]] is defined for complex argument \(z\) by
   
  +
\[
(1)\(~ ~ ~ ~ ~ ~ \mathrm{nem}_q(z)= z+z^3+q z^4\)
 
  +
\mathrm{nem}_q(z)=z+z^3+q\,z^4
  +
\tag{1}
  +
\]
   
  +
For \(q>0\), the algorithms described in the first Russian edition of «[[Суперфункции]]» (2014) cannot be applied in their original form; a small generalization is required.
  +
Thus the function \(\mathrm{nem}_q\) is treated as a [[transfer function]].
   
  +
This appears to be the last remaining attempt (as of 2025) to produce a real-holomorphic, growing transfer function whose [[superfunction]] cannot be constructed by methods already known in the literature.
For Function \(~ \mathrm{nem}_q~\) by equation (1), at \(q\!>\!0\), the algorithms, described in the first edition of the Russian version of the book «[[Суперфункции]]» (2014) cannot be applied "as is", but a small modification, generalization is required.
 
   
  +
But again, the attempt fails: both the [[superfunction]] [[SuNem]] and the [[Abel function]] [[AuNem]] are constructed using a method similar to that used for the sine function.
For this reason, function \(\mathrm{nem}_q\) by equation (1) is considered as a [[transfer function]].
 
   
  +
Associated functions:
To year 2025, the [[Nemtsov function]]
 
appears to be the last attempt to suggest a real-holomorphic growing [[transfer function]] such that its [[superfunction]] cannot be constructed with methods similar to those already described in publications.
 
   
  +
[[Inverse function]]: \(\mathrm{ArqNem}_q\), satisfying
This attempt fails. For the [[Nemtsov function]], the [[superfunction]] [[SuNem]] and the [[abelfunction]] [[AuNem]] are constructed in almost the same method used for the transfer function [[sin]].
 
  +
\[
  +
\mathrm{nem}_q(\mathrm{ArqNem}_q(z)) = z
  +
\]
   
  +
[[Superfunction]] \(\mathrm{SuNem}_q\), satisfying
In this article, properties of the [[Nemtsov function]] are described, and also some properties of the related functions:
 
  +
\[
  +
\mathrm{SuNem}_q(z+1) = \mathrm{nem}_q(\mathrm{SuNem}_q(z))
  +
\]
   
  +
[[Abel function]]: \(\mathrm{AuNem}_q=\mathrm{SuNem}_q^{-1}\), satisfying
Inverse function, denoted with [[ArqNem]],
 
  +
 
  +
\[
\(\mathrm{nem}_q(\mathrm{ArqNem}_q(z))=z\)
 
  +
\mathrm{AuNem}_q(\mathrm{nem}_q(z)) = \mathrm{AuNem}_q(z) + 1
 
  +
\]
[[Superfunction]], denoted with [[SuNem]],
 
 
\(\mathrm{SuNem}_q(z\!+\!1)=\mathrm{nem}_q\big(\mathrm{SuNem}_q(z)\big)\)
 
 
[[Abel function]], denoted with [[AuNem]],
 
 
\(\mathrm{AuNem}_q(\mathrm{nem}_q(z))=\mathrm{AuNem}_q(z)+1\)
 
 
and the corresponding [[iterate]]s
 
 
\(\mathrm{nem}_q^n(z)=\mathrm{SuNem}_q\big(n+ \mathrm{SuNem}_q(z)\big)\)
 
   
  +
Several inverse functions exist: [[ArcNem]], [[ArkNem]], and [[ArqNem]].
The inverse function is called [[ArqNem]]. This name allows to distinguish it from other inverse functions of the [[Nemtsov function]]. Two other inverse functions are called [[ArcNem]] and [[ArkNem]]. A priori, it had been difficult to guess, that namely [[ArqNem]] happens to be suitable for construction of the [[superfunction]] and the corresponding [[abelfunction]]; so, all the three versions had been assigned (designated) the different names. These inverse functions have different positions of the cuts of the range of holomorphism.
 
  +
They differ in branch-cut placement.
  +
[[ArqNem]] turns out to be the correct one for defining holomorphic non-integer iterates near the positive real axis; therefore it is adopted as the default inverse.
   
  +
== Inverse function ==
The [[Abel function]] for the Nemtov function is called [[AuNem]], to indicate, that it is constructed by the [[exotic iterates]] at the [[fixed point]] zero, that is maximal (Upper) among the fixed points of the Nemtsov function.
 
 
While no other iterates of the Nemtsov function are presented, no special mark is used to denote the iterates \(\mathrm{nem}_q^n\).
 
Later, perhaps, one additional subscript \(_{\mathrm u}\) will be
 
added to the notation, in order to distinguish this iterate from other iterates, constructed, for example, using the asymptotic behaviour of the [[Nemtsov function]] (and its iterates) at infinity.
 
 
==Inverse function==
 
   
 
<div class="thumb tright" style="float:right; margin:-4px 2px 200px 2px; width:410px; height:460px; background-color:#fff">
 
<div class="thumb tright" style="float:right; margin:-4px 2px 200px 2px; width:410px; height:460px; background-color:#fff">
 
<p style="margin:0px 0px 0px 9px;line-height:0px; width:200px; background-color:#fff">&nbsp; &nbsp; {{pic|Nembraplot.jpg|180px}}</p>
 
<p style="margin:0px 0px 0px 9px;line-height:0px; width:200px; background-color:#fff">&nbsp; &nbsp; {{pic|Nembraplot.jpg|180px}}</p>
 
<p style="margin:-440px 0px 440px 220px;line-height:0px">{{pic|Nembrant.jpg|180px}}</p>
 
<p style="margin:-440px 0px 440px 220px;line-height:0px">{{pic|Nembrant.jpg|180px}}</p>
<p style="margin:-432px 0px 4px 12px;"><small><center>\(x\!+\!\mathrm i y= \mathrm{NemBra}(q) ~\) and \(~ x\!+\!\mathrm i y= \mathrm{NemBran}(q)\) </center></small></p>
+
<p style="margin:-432px 0px 4px 12px;"><small><center>\(x+\mathrm i y=\mathrm{NemBra}(q)\) and \(x+\mathrm i y=\mathrm{NemBran}(q)\)</center></small></p>
 
</div>
 
</div>
To construct the inverse function in the complex plane, we need to identify the saddle points and choose the appropriate cut lines.<!--
 
In order to construct the inverse function in the complex plane, we need to
 
reveal the saddle points and choose the appropriate cut lines.!-->
 
For function [[ArqNem]]\(_q\), for \(q=0\) and for \(q=2\),
 
in the maps at the right hand side column of the second figure,
 
these cuts are shown with yellow lines.<br>
 
These lines connect the saddle points of the [[Nemtsov function]].<br>
 
Positions of these branch points are considered below.
 
   
  +
To construct the inverse function in the complex plane, one must locate the saddle points and choose suitable branch cuts.
For real \(q\), the [[Nemtsov function]] is real holomorphic in the whole complex plane;
 
  +
In the complex maps shown previously, the yellow lines indicate the branch cuts for functions \(\mathrm{ArqNem}_0\) and \(\mathrm{ArqNem}_2\).
  +
These lines connect the two complex branch points of \(\mathrm{ArqNem}_q\).
   
  +
For real \(q\), the Nemtsov function is real-holomorphic in the whole complex plane:
\(\mathrm{nem}_q(z^*)=\mathrm{nem}_q(z)^*\)
 
   
  +
\[
At least for positive values of the argument, the [[Nemtsov function]] grows monotonically.<br>
 
  +
\mathrm{nem}_q(z^{*}) = \mathrm{nem}_q(z)^{*}.
The monotonous growth is also property of the inverse function [[ArqNem]], that of the [[Abelfunction]] [[AuNem]] and that of the real iterates of the [[Nemtsov function]].
 
  +
\]
   
  +
For positive real arguments, the function grows monotonically.
The [[inverse function]]
 
  +
This monotonicity is inherited by its inverse \(\mathrm{ArqNem}_q\), its Abel function \(\mathrm{AuNem}_q\), and its real iterates.
\(\mathrm{ArqNem}_q\) has the [[branch point]]s. <br>
 
Two of them are complex; and one of them is expressed with function [[NemBran]].<br>
 
Parametric plot of function [[NemBran]] is shown at second picture in figure at right.<br>
 
At \(z=\mathrm{NemBran}(q)\), function \(\mathrm{ArqNem}_q(z)\) has infinite derivative.
 
   
For construction of the inverse function, important are the complex solutions \(A\) of equation \(\mathrm{nem}_q^{\prime}(A)=0\).<br>
+
The inverse function \(\mathrm{ArqNem}_q\) has complex [[branch point]]s.
One of these solutions is expressed with function [[NemBra]], id est, \(A=\mathrm{NemBra}(q)\).
+
One of them is described explicitly by the function [[NemBran]].
  +
At \(z=\mathrm{NemBran}(q)\), the derivative of \(\mathrm{ArqNem}_q(z)\) becomes infinite.
[[Parametric plot]]s of functions [[NemBra]] and [[NemBran]] are shown at right;
 
   
  +
The branch points arise from the complex solutions \(A\) of the equation
\(\mathrm{NemBran}(q)=\mathrm{nem}_q\big(\mathrm{NemBra}(q)\big)\) :
 
   
\(
+
\[
  +
\mathrm{nem}_q'(A) = 0.
\left(
 
  +
\]
\begin{array}{cc}
 
x=\mathrm{Re}\big(\mathrm{NemBra}(q)\big) \\
 
y=\mathrm{Im}\big(\mathrm{NemBra}(q)\big)
 
\end{array}\right)
 
~\), left and
 
\(~
 
\left(
 
\begin{array}{cc}
 
x=\mathrm{Re}\big(\mathrm{NemBran}(q)\big) \\
 
y=\mathrm{Im}\big(\mathrm{NemBran}(q)\big)
 
\end{array}\right)
 
~\) , right.
 
   
  +
One such solution is denoted \(\mathrm{NemBra}(q)\), and its image
For positive \(q\), both real and imaginary parts of the branch point are significantly smaller than unity.
 
  +
  +
\[
  +
\mathrm{NemBran}(q) = \mathrm{nem}_q(\mathrm{NemBra}(q)).
  +
\]
   
  +
For positive \(q\), both real and imaginary parts of \(\mathrm{NemBra}(q)\) and \(\mathrm{NemBran}(q)\) are small (well below unity).
Once the function [[NemBra]] is implemented, the efficient algorithm for evaluation of the inverse functions of the [[Nemtsov function]]
 
can be constructed.<br>
 
Three of them are denoted with symbols «[[ArcNem]]», «[[ArkNem]]», «[[ArqNem]]».
 
The last one happens to be appropriate for construction of non-integer iterates, holomorphic at least in some vicinity of the positive part of the real axis.
 
   
  +
Once \(\mathrm{NemBra}\) is implemented, an efficient algorithm for inverse functions becomes possible.
[[Complex map]] of function [[ArqNem]] are shown in the Second figure.
 
  +
Three versions—[[ArcNem]], [[ArkNem]], and [[ArqNem]]—differ only in branch-cut structure.
  +
[[ArqNem]] is best suited for constructing holomorphic non-integer iterates near the positive real axis.
   
The [[C++]] complex double implementation of function [[ArqNem]] is loaded as [[arqnem.cin]].<br>
+
The [[C++]] implementation of \(\mathrm{ArqNem}_q\) is available as [[arqnem.cin]].
In the implementation, parameter \(q\) is stored in the global variable \(Q\). Perhaps, it would not be a good solution for any software,
+
Parameter \(q\) is stored in the global variable `Q`.
  +
Before evaluating \(\mathrm{ArqNem}_q(z)\), the corresponding branch point must be computed via [[nembran.cin]] and stored in global variables (real and imaginary parts).
but this happens to be a simple way to plot pictures for book «[[Superfunctions]]» <ref name="book"/>.
 
   
  +
== Superfunction ==
Before evaluation of \(\mathrm{ArqNem}_q\) of complex argument, the complex branch point should be evaluated with routine [[nembran.cin]] and stored in the global variables tr and ti; in the version from year 2015, the real and imaginary parts of the branch point are stored as two global variables.
 
At any change of parameter \(q\), these values should be recalculated.
 
   
==Superfunction==
 
 
<div class="thumb tleft" style="float:left; margin:-4px 12px 4px -4px; background-color:#fff">
 
<div class="thumb tleft" style="float:left; margin:-4px 12px 4px -4px; background-color:#fff">
{{pic|Sunemplo4t.jpg|180px}}<small><center>\(y\!=\!\mathrm{SuNem}_{q}(x)\) for various \(q\)</center></small>
+
{{pic|Sunemplo4t.jpg|180px}}<small><center>\(y=\mathrm{SuNem}_{q}(x)\) for various \(q\)</center></small>
 
</div>
 
</div>
   
 
<div class="thumb tright" style="float:right; margin:-24px 0px 0px 2px">
 
<div class="thumb tright" style="float:right; margin:-24px 0px 0px 2px">
&nbsp; {{pic|Sunem0map6.jpg|200px}}<small><center>\(u\!+\!\mathrm i v=\mathrm{SuNem}_0(x\!+\!\mathrm i y)\)</center></small>
+
&nbsp; {{pic|Sunem0map6.jpg|200px}}<small><center>\(u+\mathrm i v=\mathrm{SuNem}_0(x+\mathrm i y)\)</center></small>
   
&nbsp; {{pic|Sunem1map6.jpg|200px}}<small><center>\(u\!+\!\mathrm i v=\mathrm{SuNem}_1(x\!+\!\mathrm i y)\)</center></small>
+
&nbsp; {{pic|Sunem1map6.jpg|200px}}<small><center>\(u+\mathrm i v=\mathrm{SuNem}_1(x+\mathrm i y)\)</center></small>
   
&nbsp; {{pic|Sunem2map6.jpg|200px}}<small><center>\(u\!+\!\mathrm i v=\mathrm{SuNem}_2(x\!+\!\mathrm i y)\)</center></small>
+
&nbsp; {{pic|Sunem2map6.jpg|200px}}<small><center>\(u+\mathrm i v=\mathrm{SuNem}_2(x+\mathrm i y)\)</center></small>
 
</div>
 
</div>
   
  +
For the transfer function \(\mathrm{nem}_q\), the [[superfunction]] \(\mathrm{SuNem}_q\) is the real-holomorphic solution \(F\) of the transfer equation
For the [[Nemtsov function]]
 
  +
\(\mathrm{nem}_q\), the [[superfunction]] \(\mathrm{SuNem}_q~\) is real-holomorphic solution \(F\) of the [[transfer equation]]
 
\[F(z\!+\!1)=\mathrm{nem}_q\big( F(z)\big)\]
 
with specific asymptotic behaviour at \(-\infty\), namely,
 
 
\[
 
\[
  +
F(z+1) = \mathrm{nem}_q(F(z)),
F(z)=\frac{1}{\sqrt{-2 z}}\left( 1-\frac{q}{\sqrt{-2 z}} + O\big( \ln(-z)/z\big) \right)\]
 
  +
\]
In order to specify function SuNem, the additional condition is assumed:
 
\[\mathrm{SuNem}_q(0)=1\]
 
(Similar condition is used to specify [[tetration]] as [[superfunction]] of exponent).
 
   
  +
with the asymptotic condition at \(-\infty\):
Explicit plot of function [[SuNem]] of the real argument is shown in figure at left, \(y\!=\!\mathrm{SuNem}_q(x)\) is plotted versus \(x\) for various values of \(q\).
 
   
  +
\[
Function [[SuNem]] grows monotonously from zero at \(-\infty\), takes value unity at zero and then grows quickly to infinity for positive values of the argument.
 
  +
F(z) = \frac{1}{\sqrt{-2z}}
The larger is parameter \(q\), the faster is the growth at \(+\infty\).
 
  +
\left( 1 - \frac{q}{\sqrt{-2z}}
This behaviour corresponds to the intuitive expectations about this function.
 
  +
+ O\!\left( \frac{\ln(-z)}{z} \right)
  +
\right),
  +
\qquad z\to -\infty.
  +
\]
  +
  +
To fully specify \(\mathrm{SuNem}_q\), an additional normalization is imposed:
  +
  +
\[
  +
\mathrm{SuNem}_q(0) = 1.
  +
\]
  +
  +
The plot \(y=\mathrm{SuNem}_q(x)\) appears at left for several values of \(q\).
  +
The function grows monotonically from \(0\) at \(-\infty\), reaches \(1\) at argument \(0\), and then increases rapidly for positive arguments.
  +
Larger \(q\) produces faster growth.
  +
  +
Complex maps \(u+\mathrm i v=\mathrm{SuNem}_q(x+\mathrm i y)\) are displayed for \(q=0\), \(q=1\), and \(q=2\).
  +
  +
To construct \(\mathrm{SuNem}_q\), one first constructs any superfunction \(F\) with the correct asymptotics, and then defines
  +
  +
\[
  +
\mathrm{SuNem}_q(z) = F(x_1 + z),
  +
\]
   
  +
where \(x_1\) is the real number satisfying \(F(x_1)=1\).
Complex maps of function [[SuNem]] are shown in figures at right for \(q\!=\!0\), \(q\!=\!1\) and for \(q\!=\!2\);
 
\[u\!+\!\mathrm i v=\mathrm{SuNem}_q(x\!+\!\mathrm i y)\]
 
Maps for different values of \(q\) look similar; however, the greater is \(q\), the faster is the growth of \(\mathrm{SuNem}_q\) along the real axis.
 
This is seen also at the explicit plot in figure at left.
 
   
  +
== Abel function ==
In order to construct function [[SuNem]], first, any superfunction \(F\) with appropriate asymptotic behaviour is constructed.
 
<br>
 
Then [[SuNem]] is set as
 
\[\mathrm{SuNem}_q(z)=F(x_1+z)\]
 
where \(x_1\) is real solution of equation \(F(x_1)=1\).
 
   
==[[Abelfunction]]==
 
 
<div style="float:left; width:200px; margin:-6px 18px 5px 0px; background-color:#fff; width:210px">
 
<div style="float:left; width:200px; margin:-6px 18px 5px 0px; background-color:#fff; width:210px">
{{pic|Aunemplot.jpg|200px}}<small><center>\(y\!=\!\mathrm{AuNem}_q(x)\) for \(q=0\),
+
{{pic|Aunemplot.jpg|200px}}<small><center>\(y=\mathrm{AuNem}_q(x)\) for \(q=0\), \(1\), \(2\)</center></small>
\(1\), \(2\).</center></small>
 
 
</div>
 
</div>
For the [[Abel function]] of the [[Nemtsov function]], notation [[AuNem]] is suggested.
 
   
  +
For the Abel function of the Nemtsov function, the notation [[AuNem]] is used.
Explicit plot \(y\!=\!\mathrm{AuNem}_q(x)\) versus \(x\) is shown in figure at left for
 
\(q\!=\!0\), \(q\!=\!1\) and \(q\!=\!2\).
 
   
  +
The explicit plot \(y=\mathrm{AuNem}_q(x)\) versus \(x\) is shown at left for \(q=0\), \(1\), and \(2\).
The same plot can be obtained, reflecting the curves for [[SuNem]] from the angle bisector of the First quadrant of the coordinate plane.
 
  +
This plot coincides with reflecting the plots of \(\mathrm{SuNem}_q\) across the bisector of the first quadrant.
  +
This symmetry provides a numerical test:
   
  +
\[
This property can be used as a numerical test of relation
 
\[\mathrm{SuNem}_q(\mathrm{АuNem}_q(x))=x\]
+
\mathrm{SuNem}_q(\mathrm{AuNem}_q(x)) = x,
  +
\qquad
it should be so, while
 
\[\mathrm{AuNem}_q=\mathrm{SuNem}_q^{-1}\]
+
\mathrm{AuNem}_q = \mathrm{SuNem}_q^{-1}.
  +
\]
As inverse of the [[superfunction]], function [[AuNem]] satisfies the [[Abel equation]]
 
\[\mathrm{AuNem}_q\big( \mathrm{nem}_q(z)\big)=\mathrm{AuNem}_q(z)+1\]
 
Function AuNem satisfies also the additional condition
 
\[\mathrm{AuNem}_q(1)=0\]
 
that is determined by the corresponding property of the SuNem, namely, that \( \mathrm{SuNem}_q(0)\!=\!1~\).
 
   
  +
As the inverse of the superfunction, the Abel function satisfies the Abel equation
The asymptotic expansion of function [[AuNem]] at zero can be found, inverting the asymptotic expansion of function [[SuNem]] at \(-\infty\).
 
The coefficients of this expansion can be found also from the [[Abel equation]].
 
   
  +
\[
==Iterates==
 
  +
\mathrm{AuNem}_q(\mathrm{nem}_q(z)) = \mathrm{AuNem}_q(z) + 1.
<div class="thumb tright" style="float:right; margin:-20px 0px 2px 8px; background-color:#fff">
 
  +
\]
{{pic|Itnem00plot.jpg|216px}}<small><center>\(y\!=\!\mathrm{nem}_0^{\,n}(x)\) versus \(x\) at various \(n\)</center></small>
 
  +
{{pic|Itnem10plot.jpg|216px}}<small><center>\(y\!=\!\mathrm{nem}_1^{\,n}(x)\) versus \(x\) at various \(n\)</center></small>
 
  +
The normalization is
{{pic|Itnem20plot.jpg|216px}}<small><center>\(y\!=\!\mathrm{nem}_2^{\,n}(x)\) versus \(x\) at various \(n\)</center></small>
 
  +
  +
\[
  +
\mathrm{AuNem}_q(1) = 0,
  +
\]
  +
  +
because \(\mathrm{SuNem}_q(0)=1\).
  +
  +
The asymptotic expansion of \(\mathrm{AuNem}_q\) near \(0\) follows from inverting the asymptotics of \(\mathrm{SuNem}_q\) at \(-\infty\).
  +
  +
== Iterates ==
  +
  +
<div class="thumb tright" style="float:right; margin:-32px 0px 2px 8px; background-color:#fff">
  +
{{pic|Itnem00plot.jpg|216px}}<small><center>\(y=\mathrm{nem}_0^{\,n}(x)\) for various \(n\)</center></small>
  +
{{pic|Itnem10plot.jpg|216px}}<small><center>\(y=\mathrm{nem}_1^{\,n}(x)\) for various \(n\)</center></small>
  +
{{pic|Itnem20plot.jpg|216px}}<small><center>\(y=\mathrm{nem}_2^{\,n}(x)\) for various \(n\)</center></small>
 
</div>
 
</div>
Using the functions [[SuNem]] and [[AuNem]], the \(n\)th [[iterate]] of the [[Nemtsov function]] can be expressed as follows:<!--
 
With functions [[SuNem]] and [[AuNem]], the \(n\)th [[iterate]] of the [[Nemtsov function]] can be expressed as follows!-->
 
\[\mathrm{nem}_q^{\,n}(z)=\mathrm{AuNem}_q\big( n+\mathrm{AuNem}_q(z)\big)\]
 
for <br>
 
\(q\!=\!0\),<br>
 
\(q\!=\!1\) and for<br>
 
\(q\!=\!2\),<br>
 
these iterates are shown in figures at right.<br>
 
These iterates look similar to other iterates of the fast growing functions.
 
   
At positive number \(n\) of iterate, the iterate shows the fast growth;
+
Using the functions \(\mathrm{SuNem}_q\) and \(\mathrm{AuNem}_q\), the \(n\)-th iterate of the Nemtsov function can be expressed as
at negative values of \(n\), the growth is slow.
 
   
  +
\[
The zeroth iterate (\(n\!=\!0\)) appears as the [[identity function]]. <br>
 
  +
\mathrm{nem}_q^{\,n}(z)
In the figures, the corresponding graphics are marked with green lines.
 
  +
= \mathrm{SuNem}_q\!\big(n + \mathrm{AuNem}_q(z)\big).
  +
\]
   
  +
Plots of the iterates for \(q=0\), \(q=1\), and \(q=2\) appear at right.
Due to the singularity of function [[ArqNem]] at zero, the non-integer iterates are not defined at zero and the negative part of the real axis, although they approach zero as the positive argument of the iterate becomes small.
 
  +
For positive integer \(n\), the iterate grows rapidly; for negative \(n\), the growth is slow.
   
  +
The zeroth iterate (\(n=0\)) is the identity function.
The curves of the iterates show the symmetry with respect to reflections from the bisector of the First quadrant of the coordinate plane, as
 
  +
In the figures, this curve appears as a green line.
  +
  +
Because \(\mathrm{ArqNem}_q\) has a singularity at \(0\), non-integer iterates are not defined at \(0\) or on the negative real axis, though they approach \(0\) from the right.
  +
<!-- Not so easy, there is cutline along the negative part of the real axis.
  +
The iterates satisfy the symmetry
  +
  +
\[
  +
\mathrm{nem}_q^{-n}(z) = \mathrm{ArqNem}_q^{\,n}(z).
  +
\]
  +
!-->
   
  +
== Applications ==
\(\mathrm{nem}_q^{-n}(z)=\mathrm{ArqNem}_q^{\,n}(z)\)
 
   
  +
The Nemtsov function was proposed as a candidate transfer function for which the superfunction and the Abel function would be difficult to construct using the [[exotic iterate]] at its fixed point \(0\).
==Applications==
 
  +
The difficulty lay mostly in constructing the correct inverse function \(\mathrm{ArqNem}_q\), choosing branch cuts appropriately, and recognizing that \(\mathrm{ArqNem}_q\) rather than \(\mathrm{ArcNem}\) or \(\mathrm{ArkNem}\) should serve as the default inverse.
The [[Nemtsov function]] had been considered as a candidate of the [[transfer function]], for which the
 
[[superfunction]] and the [[Abel function]] are difficult to construct with the [[exotic iterate]] at its [[fixed point]] zero.
 
   
  +
Other inverse functions differ only in branch-cut placement and do not yield a real-holomorphic Abel function on as wide a domain.
Indeed, the construction required certain efforts; they were related mainly with construction of the inverse function [[ArqNem]], the efficient implementation and, especially, with guessing, that namely
 
[[ArqNem]] should be declared as the default \(\mathrm{nem}_q^{-1}\) in construction of the [[Abel function]] and non-integer iterates, that are holomorphic in the most of the complex plane.
 
   
  +
Once the inverse iterates converge to the fixed point, the exotic iteration becomes straightforward.
For the [[Nemtsov function]], two other [[inverse function]]s have been constructed, they are denoted with names
 
  +
The same approach can be used for other exotic transfer functions whose expansion at the fixed point begins with a linear term and a cubic term (the cubic coefficient can always be normalized to \(1\)).
[[ArcNem]] and [[ArkNem]].
 
  +
The required transformation appears in the last row of the [[Table of superfunctions]].
<br>
 
They have different positions of the cut lines, and, at the iterates, do not provide the
 
real-holomorphic [[Abelfunction]] in the wide range as the [[ArqNem]]. In such a way, names [[ArcNem]] and [[ArkNem]] are reserved for the native inverse functions; but the [[ArqNem]]
 
is postulated as the default [[inverse function]].
 
   
  +
== Warning ==
Once the iterates of the inverse function leads to
 
the fixed point of the [[transfer function]], the [[exotic iterates]] become straightforward.
 
<br>
 
Other exotic transfer functions can be considered in the similar way, while their expansion at the fixed point begins with the identity function and the cubic term.
 
<br>
 
Without loss of generality, the coefficient at the cubic term can be treated as unity; the corresponding transform to this case is shown in the last row of the
 
[[Table of superfunctions]].
 
   
  +
The name “[[Nemtsov function]]’’ was chosen as a mnemonic connected to a notable event of the 21st century.
==Warning==
 
   
  +
It is not intended as a political statement or as an attempt to appeal to the Russian usurper or his accomplices.
The name «[[Nemtsov function]]» is chosen as a simplest mnemonic referring to the event of the 21st century.
 
  +
The ongoing rise of corruption <ref name="medvedko"/> indicates that moral appeals are meaningless in that context, even while some impostors pretend they can “[[end war in 24 hours]]’’ by [[appeasing aggression]] and [[war crime]]s.
   
  +
Nevertheless, the editor reserves the right to use the most convenient system of notation.
The naming is not an attempt to remind the Russian usurper and his crime accomplices that
 
it is not a good style, to kill the political opponent.
 
<br>
 
The growth of [[corruption]] <ref name="medvedko">
 
http://kremlin.ru/transcripts/1566 Д.Медведев. Вступительное слово на заседании Совета по противодействию коррупции. 30 сентября 2008 года, 16:25 Москва, Кремль. Д.Медведев: Коррупция в нашей стране приобрела не просто масштабный характер, она стала привычным, обыденным явлением, которое характеризует саму жизнь в нашем обществе...
 
</ref> at putin's [[pahanat]] indicates that
 
no moral criteria or concepts can be applied to the Russian usurper nor to his accomplices.
 
So, the attempt mentioned would be just vain (even through some impostors clam they can [[stop war in 24 hours]] dealing with the [[war crime]] and [[appeasing aggression]]).
 
   
  +
== Acknowledgement ==
However, Editor keeps the right to call things in the most convenient and efficient system of notations.
 
   
==Acknowledgement==
 
 
[[ChatGPT]] helped to improve this article.
 
[[ChatGPT]] helped to improve this article.
   
==References==
+
== References ==
 
{{ref}}
 
{{ref}}
   
 
{{fer}}
 
{{fer}}
   
==Keywords==
+
== Keywords ==
 
«[[Abel function]]»,
 
«[[Abel function]]»,
 
«[[ArqNem]]»,
 
«[[ArqNem]]»,

Revision as of 11:07, 7 December 2025


Nemplot.jpg
\(y=\mathrm{nem}_q(x)\)

Nem0map.jpg ArqNem0map.jpg Nem2map.jpg ArqNem2map.jpg

Complex maps: \(u+\mathrm i v=f(x+\mathrm i y)\) for \(f=\mathrm{nem}_0\), \(f=\mathrm{ArqNem}_0\) (top) and for \(f=\mathrm{nem}_2\), \(f=\mathrm{ArqNem}_2\) (bottom)

The Nemtsov function is a special kind of polynomial, suggested as an example of a transfer function in the book «Superfunctions» [1]. The description is also available as the Mizugadro Preprint [2].

The Nemtsov function \(y=\mathrm{nem}_q(x)= x+x^3+q\,x^4\) is shown in the figure at right for various \(q\ge 0\).

Complex maps of \(\mathrm{nem}_q\) are shown in the left column of the figure below. The right column shows similar maps for the inverse function \(\mathrm{ArqNem}_q=\mathrm{nem}_q^{-1}\).

This article describes the Nemtsov function and its related functions: the inverse function ArqNem, the superfunction SuNem, and the Abel function AuNem.

Motivation

The Nemtsov function serves as an example of a real-holomorphic transfer function with a real fixed point, where the usual regular iteration method for constructing a superfunction cannot be applied directly. The editor did not find any simpler example of this type other than this specific fourth-order polynomial.

The function was introduced as an attempt to construct an “exotic’’ transfer function for which a growing real-holomorphic superfunction could not be produced by the usual methods appearing in the «Table of superfunctions». This attempt failed — the superfunction for the Nemtsov function *can* be constructed, and is described below.

The expansion of the Nemtsov function at its fixed point begins with a linear term whose coefficient is unity. This prevents the use of standard regular iteration [3], which works for the exponential to base \(\sqrt{2}\).

The situation is similar to the exponential to base \(\exp(1/\mathrm e)\) [4], but the Nemtsov function lacks the quadratic term, so the exotic iteration of [4] cannot be used as is.

A closer analogy is the case of the sine function [5], but sine is antisymmetric, \(\sin(-z)=-\sin(z)\), which simplifies its analysis. The Nemtsov function for \(q>0\) has no such symmetry, leading initially to doubts whether a superfunction could be constructed.

Eventually, the construction succeeded.

The need for a special name for this function emerged on 2015-02-27, the day when Putin killed Nemtsov. As of 2025, no other scientific concept attached to that event has appeared. The total corruption in Russia [7] prevents professional investigation of the crime. Thus, the family name “Nemtsov’’ serves as a historical timestamp.

By 2025, no better notation for this polynomial has been proposed. The function symbol is written as lowercase \(\mathrm{nem}\), following the convention for mathematical functions, even though “Nemtsov’’ is a proper name. Capitalization is used only when distinguishing it from derived functions such as ArqNem, AuNem, and SuNem.

Definition and notations

Let \(q\ge 0\) be a real parameter. The Nemtsov function is defined for complex argument \(z\) by

\[ \mathrm{nem}_q(z)=z+z^3+q\,z^4 \tag{1} \]

For \(q>0\), the algorithms described in the first Russian edition of «Суперфункции» (2014) cannot be applied in their original form; a small generalization is required. Thus the function \(\mathrm{nem}_q\) is treated as a transfer function.

This appears to be the last remaining attempt (as of 2025) to produce a real-holomorphic, growing transfer function whose superfunction cannot be constructed by methods already known in the literature.

But again, the attempt fails: both the superfunction SuNem and the Abel function AuNem are constructed using a method similar to that used for the sine function.

Associated functions:

Inverse function: \(\mathrm{ArqNem}_q\), satisfying \[ \mathrm{nem}_q(\mathrm{ArqNem}_q(z)) = z \]

Superfunction \(\mathrm{SuNem}_q\), satisfying \[ \mathrm{SuNem}_q(z+1) = \mathrm{nem}_q(\mathrm{SuNem}_q(z)) \]

Abel function: \(\mathrm{AuNem}_q=\mathrm{SuNem}_q^{-1}\), satisfying

\[ \mathrm{AuNem}_q(\mathrm{nem}_q(z)) = \mathrm{AuNem}_q(z) + 1 \]

Several inverse functions exist: ArcNem, ArkNem, and ArqNem. They differ in branch-cut placement. ArqNem turns out to be the correct one for defining holomorphic non-integer iterates near the positive real axis; therefore it is adopted as the default inverse.

Inverse function

    Nembraplot.jpg

Nembrant.jpg

\(x+\mathrm i y=\mathrm{NemBra}(q)\) and \(x+\mathrm i y=\mathrm{NemBran}(q)\)

To construct the inverse function in the complex plane, one must locate the saddle points and choose suitable branch cuts. In the complex maps shown previously, the yellow lines indicate the branch cuts for functions \(\mathrm{ArqNem}_0\) and \(\mathrm{ArqNem}_2\). These lines connect the two complex branch points of \(\mathrm{ArqNem}_q\).

For real \(q\), the Nemtsov function is real-holomorphic in the whole complex plane:

\[ \mathrm{nem}_q(z^{*}) = \mathrm{nem}_q(z)^{*}. \]

For positive real arguments, the function grows monotonically. This monotonicity is inherited by its inverse \(\mathrm{ArqNem}_q\), its Abel function \(\mathrm{AuNem}_q\), and its real iterates.

The inverse function \(\mathrm{ArqNem}_q\) has complex branch points. One of them is described explicitly by the function NemBran. At \(z=\mathrm{NemBran}(q)\), the derivative of \(\mathrm{ArqNem}_q(z)\) becomes infinite.

The branch points arise from the complex solutions \(A\) of the equation

\[ \mathrm{nem}_q'(A) = 0. \]

One such solution is denoted \(\mathrm{NemBra}(q)\), and its image

\[ \mathrm{NemBran}(q) = \mathrm{nem}_q(\mathrm{NemBra}(q)). \]

For positive \(q\), both real and imaginary parts of \(\mathrm{NemBra}(q)\) and \(\mathrm{NemBran}(q)\) are small (well below unity).

Once \(\mathrm{NemBra}\) is implemented, an efficient algorithm for inverse functions becomes possible. Three versions—ArcNem, ArkNem, and ArqNem—differ only in branch-cut structure. ArqNem is best suited for constructing holomorphic non-integer iterates near the positive real axis.

The C++ implementation of \(\mathrm{ArqNem}_q\) is available as arqnem.cin. Parameter \(q\) is stored in the global variable `Q`. Before evaluating \(\mathrm{ArqNem}_q(z)\), the corresponding branch point must be computed via nembran.cin and stored in global variables (real and imaginary parts).

Superfunction

Sunemplo4t.jpg
\(y=\mathrm{SuNem}_{q}(x)\) for various \(q\)
  Sunem0map6.jpg
\(u+\mathrm i v=\mathrm{SuNem}_0(x+\mathrm i y)\)
  Sunem1map6.jpg
\(u+\mathrm i v=\mathrm{SuNem}_1(x+\mathrm i y)\)
  Sunem2map6.jpg
\(u+\mathrm i v=\mathrm{SuNem}_2(x+\mathrm i y)\)

For the transfer function \(\mathrm{nem}_q\), the superfunction \(\mathrm{SuNem}_q\) is the real-holomorphic solution \(F\) of the transfer equation

\[ F(z+1) = \mathrm{nem}_q(F(z)), \]

with the asymptotic condition at \(-\infty\):

\[ F(z) = \frac{1}{\sqrt{-2z}} \left( 1 - \frac{q}{\sqrt{-2z}} + O\!\left( \frac{\ln(-z)}{z} \right) \right), \qquad z\to -\infty. \]

To fully specify \(\mathrm{SuNem}_q\), an additional normalization is imposed:

\[ \mathrm{SuNem}_q(0) = 1. \]

The plot \(y=\mathrm{SuNem}_q(x)\) appears at left for several values of \(q\). The function grows monotonically from \(0\) at \(-\infty\), reaches \(1\) at argument \(0\), and then increases rapidly for positive arguments. Larger \(q\) produces faster growth.

Complex maps \(u+\mathrm i v=\mathrm{SuNem}_q(x+\mathrm i y)\) are displayed for \(q=0\), \(q=1\), and \(q=2\).

To construct \(\mathrm{SuNem}_q\), one first constructs any superfunction \(F\) with the correct asymptotics, and then defines

\[ \mathrm{SuNem}_q(z) = F(x_1 + z), \]

where \(x_1\) is the real number satisfying \(F(x_1)=1\).

Abel function

Aunemplot.jpg
\(y=\mathrm{AuNem}_q(x)\) for \(q=0\), \(1\), \(2\)

For the Abel function of the Nemtsov function, the notation AuNem is used.

The explicit plot \(y=\mathrm{AuNem}_q(x)\) versus \(x\) is shown at left for \(q=0\), \(1\), and \(2\). This plot coincides with reflecting the plots of \(\mathrm{SuNem}_q\) across the bisector of the first quadrant. This symmetry provides a numerical test:

\[ \mathrm{SuNem}_q(\mathrm{AuNem}_q(x)) = x, \qquad \mathrm{AuNem}_q = \mathrm{SuNem}_q^{-1}. \]

As the inverse of the superfunction, the Abel function satisfies the Abel equation

\[ \mathrm{AuNem}_q(\mathrm{nem}_q(z)) = \mathrm{AuNem}_q(z) + 1. \]

The normalization is

\[ \mathrm{AuNem}_q(1) = 0, \]

because \(\mathrm{SuNem}_q(0)=1\).

The asymptotic expansion of \(\mathrm{AuNem}_q\) near \(0\) follows from inverting the asymptotics of \(\mathrm{SuNem}_q\) at \(-\infty\).

Iterates

Itnem00plot.jpg
\(y=\mathrm{nem}_0^{\,n}(x)\) for various \(n\)
Itnem10plot.jpg
\(y=\mathrm{nem}_1^{\,n}(x)\) for various \(n\)
Itnem20plot.jpg
\(y=\mathrm{nem}_2^{\,n}(x)\) for various \(n\)

Using the functions \(\mathrm{SuNem}_q\) and \(\mathrm{AuNem}_q\), the \(n\)-th iterate of the Nemtsov function can be expressed as

\[ \mathrm{nem}_q^{\,n}(z) = \mathrm{SuNem}_q\!\big(n + \mathrm{AuNem}_q(z)\big). \]

Plots of the iterates for \(q=0\), \(q=1\), and \(q=2\) appear at right. For positive integer \(n\), the iterate grows rapidly; for negative \(n\), the growth is slow.

The zeroth iterate (\(n=0\)) is the identity function. In the figures, this curve appears as a green line.

Because \(\mathrm{ArqNem}_q\) has a singularity at \(0\), non-integer iterates are not defined at \(0\) or on the negative real axis, though they approach \(0\) from the right.

Applications

The Nemtsov function was proposed as a candidate transfer function for which the superfunction and the Abel function would be difficult to construct using the exotic iterate at its fixed point \(0\). The difficulty lay mostly in constructing the correct inverse function \(\mathrm{ArqNem}_q\), choosing branch cuts appropriately, and recognizing that \(\mathrm{ArqNem}_q\) rather than \(\mathrm{ArcNem}\) or \(\mathrm{ArkNem}\) should serve as the default inverse.

Other inverse functions differ only in branch-cut placement and do not yield a real-holomorphic Abel function on as wide a domain.

Once the inverse iterates converge to the fixed point, the exotic iteration becomes straightforward. The same approach can be used for other exotic transfer functions whose expansion at the fixed point begins with a linear term and a cubic term (the cubic coefficient can always be normalized to \(1\)). The required transformation appears in the last row of the Table of superfunctions.

Warning

The name “Nemtsov function’’ was chosen as a mnemonic connected to a notable event of the 21st century.

It is not intended as a political statement or as an attempt to appeal to the Russian usurper or his accomplices. The ongoing rise of corruption [7] indicates that moral appeals are meaningless in that context, even while some impostors pretend they can “end war in 24 hours’’ by appeasing aggression and war crimes.

Nevertheless, the editor reserves the right to use the most convenient system of notation.

Acknowledgement

ChatGPT helped to improve this article.

References

  1. https://www.amazon.co.jp/-/en/Dmitrii-Kouznetsov/dp/6202672862
    https://www.morebooks.de/shop-ui/shop/product/978-620-2-67286-3
    https://mizugadro.mydns.jp/BOOK/468.pdf
    Dmitrii Kouznetsov. Superfunctions. Lambert Academic Publishing, 2020.
  2. https://mizugadro.mydns.jp/PAPERS/2016nemtsov.pdf
    Dmitrii Kouznetsov. Nemtsov function and its iterates. Mizugadro Preprint, 2016.
  3. http://www.ams.org/journals/mcom/2010-79-271/S0025-5718-10-02342-2/home.html
    https://mizugadro.mydns.jp/PAPERS/2010sqrt2.pdf
    D. Kouznetsov, H. Trappmann. Portrait of the four regular super-exponentials to base sqrt(2). Mathematics of Computation, 2010, v.79, p.1727–1756.
  4. 4.0 4.1 http://www.ams.org/journals/mcom/0000-000-00/S0025-5718-2012-02590-7/S0025-5718-2012-02590-7.pdf
    https://mizugadro.mydns.jp/PAPERS/2012e1eMcom2590.pdf
    H. Trappmann, D. Kouznetsov. Computation of the Two Regular Super-Exponentials to base exp(1/e). Mathematics of Computation, 81 (2012), 2207–2227.
  5. http://www.pphmj.com/references/8246.htm
    https://mizugadro.mydns.jp/PAPERS/2014susin.pdf
    Dmitrii Kouznetsov. SUPER SIN. Far East Journal of Mathematical Sciences 85(2), 219–238 (2014).
  6. http://nemtsov.ru Борис Немцов
  7. 7.0 7.1 http://kremlin.ru/transcripts/1566
    D. Medvedev: Corruption in our country has acquired not just a large-scale character; it has become a habitual, everyday phenomenon... (2008)