Document 10457433

advertisement
Hindawi Publishing Corporation
International Journal of Mathematics and Mathematical Sciences
Volume 2013, Article ID 929475, 5 pages
http://dx.doi.org/10.1155/2013/929475
Research Article
On the Set of Fixed Points and Periodic Points of
Continuously Differentiable Functions
Aliasghar Alikhani-Koopaei
Department of Mathematics, Berks College, Pennsylvania State University, Tulpehocken Road, P.O. Box 7009,
Reading, PA 19610-6009, USA
Correspondence should be addressed to Aliasghar Alikhani-Koopaei; axa12@psu.edu
Received 6 December 2012; Accepted 16 February 2013
Academic Editor: Paolo Ricci
Copyright © 2013 Aliasghar Alikhani-Koopaei. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
cited.
In recent years, researchers have studied the size of different sets related to the dynamics of self-maps of an interval. In this note
we investigate the sets of fixed points and periodic points of continuously differentiable functions and show that typically such
functions have a finite set of fixed points and a countable set of periodic points.
1. Introduction and Notation
The set of periodic points of self-maps of intervals has been
studied for different reasons. The functions with smaller sets
of periodic points are more likely not to share a periodic
point. Of course, one has to decide what “big” or “small”
means and how to describe this notion. In this direction one
would be interested in studying the size of the sets of periodic
points of self-maps of an interval, in particular, and other
sets arising in dynamical systems in general (see [1–5]). For
example, typically continuous functions have a first category
set of periodic points (see [1, 5]). This result was generalized
in [2] for the set of chain recurrent points. At times, even
the smallness of these sets in some sense could be useful. For
example, in [6] we showed that two commuting continuous
self-maps of an interval share a periodic point if one has a
countable set of periodic points. Schwartz (see [7]) was able to
show that if one of the two commuting continuous functions
is also continuously differentiable, then it would necessarily
follow that the functions share a periodic point. Schwartz’s
result along with the results given in [6] may suggest that
continuously differentiable functions have a countable set of
periodic points. This is not true in general. However, in this
note we show that typically such functions have a finite set
of fixed points and a countable set of periodic points. Here
𝐹1 (𝑓) = {π‘₯ : 𝑓(π‘₯) = π‘₯} denotes the set of fixed points of 𝑓.
For 𝑓 and 𝑛 ∈ N we define 𝑓𝑛 (π‘₯) by induction:
𝑓𝑛 (π‘₯) = 𝑓 (𝑓𝑛−1 (π‘₯))
with 𝑓0 (π‘₯) = π‘₯,
𝑓1 (π‘₯) = 𝑓 (π‘₯) .
(1)
The orbit of π‘₯ under 𝑓 is given by the sequence orb(π‘₯, 𝑓) =
𝑛
𝑛
{𝑓𝑖 (π‘₯)}∞
𝑖=0 . For 𝑛 ∈ N, let 𝐹𝑛 (𝑓) = {π‘₯ : 𝑓 (π‘₯) = π‘₯} and 𝑃𝑓 be
the set of periodic points of order 𝑛; that is,
𝑃𝑓𝑛 = 𝐹𝑛 (𝑓) \ ∪π‘˜<𝑛 πΉπ‘˜ (𝑓) .
(2)
Two functions on a given interval are said to be of the same
monotone type if both are either strictly increasing or strictly
decreasing on that interval. Here, for a partition 𝑇 of the
interval [π‘Ž, 𝑏], ‖𝑇‖ is the length of the largest subinterval of
𝑇, 𝐡(𝑓, πœ–) is the open ball about 𝑓 with radius πœ–, 𝐴∘ denotes
the set of interior points of 𝐴, and πœ†(𝐼) is the length of the
interval 𝐼.
2. Continuously Differentiable Functions
For 𝐼 = [π‘Ž, 𝑏], consider 𝐢(𝐼, 𝐼) and 𝐢1 (𝐼, 𝐼) to be the family
of all continuous maps and continuously differentiable maps
2
International Journal of Mathematics and Mathematical Sciences
from 𝐼 into itself, respectively. Recall that the usual metrics 𝜌0
and 𝜌1 on 𝐢(𝐼, 𝐼) and 𝐢1 (𝐼, 𝐼), respectively, are given by
󡄨
󡄨
𝜌0 (𝑓, 𝑔) = sup 󡄨󡄨󡄨𝑓 (π‘₯) − 𝑔 (π‘₯)󡄨󡄨󡄨 for 𝑓, 𝑔 ∈ 𝐢 (𝐼, 𝐼) ,
π‘₯∈𝐼
σΈ€ 
σΈ€ 
(3)
1
𝜌1 (𝑓, 𝑔) = 𝜌0 (𝑓, 𝑔) + 𝜌0 (𝑓 , 𝑔 )
for 𝑓, 𝑔 ∈ 𝐢 (𝐼, 𝐼) .
It is well known that the metric spaces (𝐢(𝐼, 𝐼); 𝜌0 ) and
(𝐢1 (𝐼, 𝐼); 𝜌1 ) are complete and hence Baire’s category theorem
holds in these spaces. We say that a typical function in 𝐢(𝐼, 𝐼)
or 𝐢1 (𝐼, 𝐼) has a certain property if the set of those functions
which does not have this property is of first category in 𝐢(𝐼, 𝐼)
or in 𝐢1 (𝐼, 𝐼). (Some authors prefer using the term generic
instead of typical.) It is known that typically continuous
self-maps of an interval have 𝜎-perfect, measure zero sets
of periodic points (see [2]). Here we show that typically
members of (𝐢1 (𝐼, 𝐼), 𝜌1 ) have a finite set of fixed points and
a countable set of periodic points.
1
Lemma 1. Let 𝑛 ≥ 1 and 𝑓 ∈ 𝐢 (𝐼, 𝐼) so that 𝐹𝑛 (𝑓) is not
finite. Then there exists π‘₯0 ∈ 𝐼 so that 𝑓𝑛 (π‘₯0 ) = π‘₯0 and
(𝑓𝑛 )σΈ€  (π‘₯0 ) = 1.
Proof. Suppose 𝐹𝑛 (𝑓) is not finite, and then we can choose
a strictly monotone sequence in 𝐹𝑛 (𝑓). Without loss of
generality, assume {π‘₯𝑖 }∞
𝑖=1 ⊂ 𝐹𝑛 (𝑓) so that π‘₯𝑖 < π‘₯𝑖+1 for
each 𝑖 ≥ 1. Let β„Ž(π‘₯) = 𝑓𝑛 (π‘₯) − π‘₯, and then, for each 𝑖,
β„Ž(π‘₯𝑖 ) = β„Ž(π‘₯𝑖+1 ) = 0. Thus by Rolle’s Theorem we have 𝑦𝑖 ,
π‘₯𝑖 < 𝑦𝑖 < π‘₯𝑖+1 so that β„ŽσΈ€  (𝑦𝑖 ) = 0. Let π‘₯0 = lim𝑖 → ∞ π‘₯𝑖 , then
π‘₯0 ∈ 𝐼, and lim𝑖 → ∞ 𝑦𝑖 = π‘₯0 , and from the continuity of 𝑓𝑛
and (𝑓𝑛 )σΈ€  it follows that 𝑓𝑛 (π‘₯0 ) = π‘₯0 and (𝑓𝑛 )σΈ€  (π‘₯0 ) = 1.
Lemma 2. For each 𝑓 ∈ 𝐢1 (𝐼, 𝐼) and 0 < πœ– < 1 there is a
polynomial 𝑝 : 𝐼 → πΌπ‘œ with 𝜌1 (𝑓, 𝑝) < πœ–.
Proof. Let 𝑀 = supπ‘₯∈𝐼 {|𝑓󸀠 (π‘₯)|, |π‘Ž|, |𝑏|, 1} where 𝐼 = [π‘Ž, 𝑏].
Take 𝛽 = πœ–/8𝑀, 𝛼 = 𝛽(π‘Ž + 𝑏)/2, and 𝑔(π‘₯) = 𝛼 + (1 − 𝛽)𝑓(π‘₯).
It is easy to see that 0 < 𝛽 < 1, 𝑔(𝐼) ⊂ πΌπ‘œ , and 𝜌1 (𝑓, 𝑔) < πœ–/4.
Let
πœ–1 =
πœ–
1
πœ–
, }
inf {𝑔 (π‘₯) − π‘Ž, 𝑏 − 𝑔 (π‘₯) ,
2 π‘₯∈[π‘Ž,𝑏]
8 (𝑏 − π‘Ž) 8
(4)
and 𝑠(𝑑) be a polynomial with 𝜌0 (𝑔󸀠 , 𝑠) < πœ–1 /(𝑏 − π‘Ž + 1). Then
π‘₯
𝑝(π‘₯) = 𝑔(π‘Ž) + ∫π‘Ž 𝑠(𝑑)𝑑𝑑 is a polynomial and 𝑔(π‘₯) − 𝑝(π‘₯) =
π‘₯
∫π‘Ž (𝑔󸀠 (𝑑) − 𝑠(𝑑))𝑑𝑑. Thus we have
𝜌1 (𝑔, 𝑝) = 𝜌0 (𝑔󸀠 , 𝑠) + 𝜌0 (𝑔, 𝑝) ≤
πœ–1
πœ– (𝑏 − π‘Ž)
+ 1
(𝑏 − π‘Ž) + 1 (𝑏 − π‘Ž) + 1
πœ–
= πœ–1 < ,
4
(5)
hence 𝜌1 (𝑓, 𝑝) ≤ 𝜌1 (𝑓, 𝑔) + 𝜌1 (𝑔, 𝑝) < πœ–/2. From 𝑔(𝐼) ⊂ [π‘Ž +
2πœ–1 , 𝑏 − 2πœ–1 ] and 𝜌1 (𝑔, 𝑝) ≤ πœ–1 it follows that 𝑝(𝐼) ⊂ πΌπ‘œ .
1
Lemma 3. Let 𝑛 > 2 be a positive integer and 𝑓 ∈ 𝐢 (𝐼, 𝐼).
If π‘₯0 is a periodic point of order 𝑛 with (𝑓𝑛 )σΈ€  (π‘₯0 ) = 1, then
orb(π‘₯0 , 𝑓) ∩ {π‘Ž, 𝑏} = 0.
Proof. We show that π‘Ž ∉ orb(π‘₯0 , 𝑓). The case 𝑏 ∉ orb(π‘₯0 , 𝑓)
follows similarly. Let orb(π‘₯0 , 𝑓) = {π‘Ž = π‘₯0 , π‘₯1 , . . . , π‘₯𝑛 = π‘₯0 }
with π‘₯𝑖 = 𝑓(π‘₯𝑖−1 ) for 1 ≤ 𝑖 ≤ 𝑛. Since 𝑛 > 2, there exist
π‘₯𝑖 ∈ orb(π‘₯0 , 𝑓) ∩ (π‘Ž, 𝑏) and 1 ≤ 𝑠 ≤ 𝑛 − 1 so that 𝑓𝑠 (π‘₯𝑖 ) =
π‘Ž. Since 𝑓𝑠 ∈ 𝐢1 (𝐼, 𝐼) and (𝑓𝑠 )σΈ€  (π‘₯0 ) =ΜΈ 0, there exists 𝛿1 > 0
so that π‘Ž < π‘₯𝑖 − 𝛿1 < π‘₯𝑖 < π‘₯𝑖 + 𝛿1 < 𝑏 and 𝑓𝑠 is strictly
increasing or strictly decreasing on 𝐽 = [π‘₯𝑖 − 𝛿1 , π‘₯𝑖 + 𝛿1 ]. Let
𝑓𝑠 be strictly increasing on 𝐽, and then for π‘₯𝑖 − 𝛿1 ≤ π‘₯ ≤ π‘₯𝑖 we
have 𝑓𝑠 (π‘₯) ≤ 𝑓𝑠 (π‘₯𝑖 ) = π‘Ž; hence, 𝑓𝑠 |[π‘₯𝑖 −𝛿1 ,π‘₯𝑖 ] = π‘Ž, implying that
(𝑓𝑠 )σΈ€  (π‘₯𝑖 ) = 0, a contradiction to (𝑓𝑛 )σΈ€  (π‘₯0 ) = 1. On the other
hand when 𝑓𝑠 is strictly decreasing on 𝐽, for π‘₯𝑖 ≤ π‘₯ ≤ π‘₯𝑖 + 𝛿1
we have 𝑓𝑠 (π‘₯) ≤ 𝑓𝑠 (π‘₯𝑖 ) = π‘Ž; hence, 𝑓𝑠 |[π‘₯𝑖 ,π‘₯𝑖 +𝛿1 ] = π‘Ž implying
that (𝑓𝑠 )σΈ€  (π‘₯𝑖 ) = 0, a contradiction to (𝑓𝑛 )σΈ€  (π‘₯0 ) = 1.
Lemma 4. For π‘˜ ≥ 1, the set
σΈ€ 
Eπ‘˜ = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : πΉπ‘˜ (𝑓) ∩ {π‘₯ : (π‘“π‘˜ ) (π‘₯) = 1} =ΜΈ 0}
(6)
is closed in 𝐢1 (𝐼, 𝐼).
Proof. Let 𝑔𝑛 ∈ Eπ‘˜ and 𝜌1 (𝑔𝑛 , 𝑔) → 0. Then there exists
π‘˜
π‘˜ σΈ€ 
{𝑑𝑛 }∞
𝑛=1 ⊂ 𝐼 such that 𝑔𝑛 (𝑑𝑛 ) = 𝑑𝑛 and (𝑔𝑛 ) (𝑑𝑛 ) = 1. Without
loss of generality, we may assume that lim𝑛 → ∞ 𝑑𝑛 = 𝑑0 , then
𝑑0 ∈ 𝐼. Let πœ– > 0 be arbitrary. Due to the uniform continuity of
π‘”π‘˜ on 𝐼 there exists a positive integer 𝑛1 such that, for 𝑛 ≥ 𝑛1 ,
|π‘”π‘˜ (𝑑𝑛 ) − π‘”π‘˜ (𝑑0 )| < πœ–. Since 𝜌0 (𝑔𝑛 , 𝑔) → 0, there exists a
positive integer 𝑛2 such that, for 𝑛 ≥ 𝑛2 , 𝜌0 (π‘”π‘›π‘˜ , π‘”π‘˜ ) < πœ–.
Choose the positive integer 𝑛3 so that |𝑑𝑛 − 𝑑0 | < πœ– for 𝑛 ≥ 𝑛3 ,
and let π‘š1 = max(𝑛1 , 𝑛2 , 𝑛3 ). Then for π‘š ≥ π‘š1 we have
󡄨󡄨 π‘˜
󡄨
󡄨󡄨𝑔 (𝑑0 ) − 𝑑0 󡄨󡄨󡄨
󡄨
󡄨
󡄨󡄨 π‘˜
󡄨 󡄨
󡄨
π‘˜
≤ 󡄨󡄨󡄨𝑔 (𝑑0 ) − π‘”π‘˜ (π‘‘π‘š )󡄨󡄨󡄨󡄨 + σ΅„¨σ΅„¨σ΅„¨σ΅„¨π‘”π‘˜ (π‘‘π‘š ) − π‘”π‘š
(π‘‘π‘š )󡄨󡄨󡄨󡄨
󡄨 π‘˜
󡄨
+ σ΅„¨σ΅„¨σ΅„¨σ΅„¨π‘”π‘š
(π‘‘π‘š ) − 𝑑0 󡄨󡄨󡄨󡄨
󡄨
󡄨
π‘˜
≤ πœ– + 𝜌0 (π‘”π‘˜ , π‘”π‘š
) + σ΅„¨σ΅„¨σ΅„¨π‘‘π‘š − 𝑑0 󡄨󡄨󡄨 ≤ 3πœ–,
(7)
Implying that π‘”π‘˜ (𝑑0 ) = 𝑑0 . We also have
󡄨󡄨 π‘˜ σΈ€ 
󡄨
󡄨󡄨(𝑔 ) (𝑑𝑛 ) − (π‘”π‘˜ )σΈ€  (𝑑𝑛 )󡄨󡄨󡄨 ≤ lim 𝜌1 (π‘”π‘˜ , π‘”π‘˜ ) = 0.
𝑛
󡄨󡄨 𝑛
󡄨󡄨 𝑛 → ∞
(8)
Thus we have lim𝑛 → ∞ |1 − (π‘”π‘˜ )σΈ€  (𝑑𝑛 )| = |1 − (π‘”π‘˜ )σΈ€  (𝑑0 )| = 0 and
𝑔 ∈ Eπ‘˜ .
Theorem 5. There exists a residual subset 𝑀1 of 𝐢1 (𝐼, 𝐼) such
that, for every 𝑓 ∈ 𝑀1 , 𝐹1 (𝑓) is finite.
Proof. If 𝐹1 (𝑓) has infinitely many elements, then from
Lemma 1 it follows that there exists a point π‘₯0 ∈ 𝐹1 (𝑓) so
that 𝑓󸀠 (π‘₯0 ) = 1. Let
E1 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : ∃π‘₯0 ∈ 𝐼 ∩ 𝐹1 (𝑓) , 𝑓󸀠 (π‘₯0 ) = 1} . (9)
From Lemma 4 the set E1 is closed in 𝐢1 (𝐼, 𝐼). To show that
E1 has no interior point, let 𝑓 ∈ E1 and πœ– > 0, and then, by
International Journal of Mathematics and Mathematical Sciences
Lemma 2, there exists a polynomial 𝑝 such that 𝜌1 (𝑓, 𝑝) < πœ–/2
and 𝑝(𝐼) ⊂ [π‘Ž + πœ–1 , 𝑏 − πœ–1 ] for some 0 < πœ–1 < πœ–/2. Let
𝑀 (𝑝) = {π‘₯ : 𝑝󸀠 (π‘₯) = 1} = {𝑑1 , 𝑑2 , . . . , π‘‘π‘š } .
(10)
Choose 0 < πœ–2 < πœ–1 so that 𝑝(π‘₯) − π‘₯ =ΜΈ πœ–2 for all π‘₯ ∈ 𝑀(𝑝) and
take β„Ž = 𝑝 − πœ–2 . Then β„Ž ∈ 𝐢1 (𝐼, 𝐼), 𝜌1 (β„Ž, 𝑓) < πœ–, 𝑀(β„Ž) = {π‘₯ :
β„ŽσΈ€  (π‘₯) = 1} = 𝑀(𝑝), and β„Ž(π‘₯) =ΜΈ π‘₯ on 𝑀(β„Ž), implying that E1
is of first category and 𝑀1 = 𝐢1 (𝐼, 𝐼) \ E1 is residual.
Theorem 6. The set of functions 𝑓 ∈ 𝐢1 (𝐼, 𝐼) with 𝑓(π‘Ž) =
𝑏, 𝑓(𝑏) = π‘Ž, and (𝑓2 )σΈ€  (π‘Ž) = 1 is a nowhere dense subset of
𝐢1 (𝐼, 𝐼).
Proof. Let 𝐴 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : 𝑓(π‘Ž) = 𝑏, 𝑓(𝑏) =
π‘Ž, and (𝑓2 )σΈ€  (π‘Ž) = 1}. It is easy to see that 𝐴 is a closed subset
of 𝐢1 (𝐼, 𝐼). To show that 𝐴 is nowhere dense, let 𝑓 ∈ 𝐢1 (𝐼, 𝐼)
and πœ– > 0. Choose 0 < 𝛾 < min{πœ–/(4(𝑏 − π‘Ž) + 4), 1} and
π‘Ž < 𝑐 < 𝑏 such that |𝑓(𝑑) − 𝑓(π‘Ž)| < (𝑏 − π‘Ž)[1 − πœ–/4] for
π‘₯
𝑑 ∈ [π‘Ž, 𝑐]. Let β„Ž(π‘₯) = 𝑓(π‘₯) + ∫π‘Ž π‘˜(𝑑)𝑑𝑑, where
−𝛾,
{
{
{
{
{
{
{
0,
{
{
{
{
π‘˜ (𝑑) = { 𝛾
,
{
{
{
3
{
{
{
{
{
{
{0,
{linear,
𝑑 = π‘Ž,
𝑐−π‘Ž
,
4
(𝑐 − π‘Ž)
𝑑=π‘Ž+
,
2
𝑑 = 𝑐, 𝑑 = 𝑏,
otherwise.
𝑑=π‘Ž+
(11)
It is clear that β„Ž(π‘Ž) = 𝑓(π‘Ž) = 𝑏, β„Ž(𝑏) = 𝑓(𝑏) = π‘Ž, β„ŽσΈ€  (π‘Ž) =
𝑓 (π‘Ž) − 𝛾, and β„ŽσΈ€  (𝑏) = 𝑓󸀠 (𝑏). It is easy to see that β„Ž ∈ 𝐢1 (𝐼, 𝐼),
(β„Ž2 )σΈ€  (π‘Ž) = (β„Ž2 )σΈ€  (𝑏) =ΜΈ 1, and 𝜌1 (𝑓, β„Ž) < πœ–/2.
σΈ€ 
Theorem 7. Let πœ– > 0, 𝑓 ∈ 𝐢1 (𝐼, 𝐼), and 𝐴 𝑛 = {π‘₯ : (𝑓𝑛 )σΈ€  (π‘₯) =
1} ∩ 𝐹𝑛 (𝑓) be a finite set, and π‘œπ‘Ÿπ‘(π‘₯, 𝑓) ⊂ (π‘Ž, 𝑏) for π‘₯ ∈ 𝐴 𝑛 .
Then there exists a function 𝑔 ∈ 𝐢1 (𝐼, 𝐼) with 𝜌1 (𝑓, 𝑔) < πœ–/4
such that {π‘₯ : (𝑔𝑛 )σΈ€  (π‘₯) = 1} ∩ 𝐹𝑛 (𝑔) = 0.
Proof. Let 𝐴 𝑛 = {π‘₯ : (𝑓𝑛 )σΈ€  (π‘₯) = 1} ∩ 𝐹𝑛 (𝑓) = {𝑧𝑗 }𝑙𝑗=1 . Let
𝐡 = {𝑑𝑗 }π‘˜π‘—=1 be the elements of 𝐴 𝑛 with distinct orbits, that is,
orb(𝑑𝑖 , 𝑓) ∩ orb(𝑑𝑗 , 𝑓) = 0 for 𝑖 =ΜΈ 𝑗 and ∪π‘˜π‘—=1 orb(𝑑𝑗 , 𝑓) ⊇ 𝐴 𝑛 .
It is clear that each 𝑑𝑗 ∈ 𝐡 is a periodic point of 𝑓 with some
period π‘š where π‘š is a factor of 𝑛. This suggests that if one
can construct a function 𝑔 that is sufficiently close to 𝑓, and
either 𝑑𝑗 ∉ π‘ƒπ‘“π‘š or (π‘“π‘š )σΈ€  (𝑑𝑗 ) =ΜΈ 1, then
𝑛 σΈ€ 
{π‘₯ : (𝑔 ) (π‘₯) = 1} ∩ 𝐹𝑛 (𝑔) ⊆ 𝐴 𝑛 \ orb (𝑑𝑗 , 𝑓) .
(12)
By repeating this process for each 𝑑𝑗 ∈ 𝐡 in a finite number
of steps we can construct the desired function 𝑔. Thus for
convenience, we assume that, for 1 ≤ 𝑗 ≤ π‘˜, 𝑧𝑗 ∈ 𝑃𝑓𝑛 .
Consider the partition 𝑇 of [π‘Ž, 𝑏] obtained by {𝑓𝑖 (𝑑𝑗 ), 1 ≤
𝑖 ≤ 𝑛, 1 ≤ 𝑗 ≤ π‘˜}, πœ–1 = (1/4) min{πœ–, β€– 𝑇 β€–}, and choose a
positive 𝛿 less than πœ–1 such that, for each 𝑖, 1 ≤ 𝑖 ≤ 𝑛 and each
𝑗, 1 ≤ 𝑗 ≤ π‘˜, and if |𝑑−𝑑𝑗 | < 𝛿, then |𝑓𝑖 (𝑑)−𝑓𝑖 (𝑑𝑗 )| < πœ–1 . Given
that, for π‘₯ ∈ 𝐴 𝑛 , orb(π‘₯, 𝑓) ⊂ (π‘Ž, 𝑏), (𝑓𝑛 )σΈ€  is continuous, and
(𝑓𝑛 )σΈ€  (π‘₯) = 1 on 𝐡, we may choose the nondegenerate closed
3
intervals 𝐻𝑗 , 1 ≤ 𝑗 ≤ π‘˜ of length less than 𝛿 and the positive
numbers 𝛾𝑗 such that
(i) 𝑑𝑗 ∈ (𝐻𝑗 )∘ is the midpoint of 𝐻𝑗 and 𝑓𝑖 (𝐻𝑗 ) ∩
orb(𝑑𝑗 , 𝑓) = {𝑓𝑖 (𝑑𝑗 )} for 0 ≤ 𝑖 ≤ 𝑛,
(ii) 𝑓 is strictly monotone on each 𝐻𝑗,𝑖 = 𝑓𝑖 (𝐻𝑗 ) for 0 ≤
𝑖 ≤ 𝑛 − 1, 1 ≤ 𝑗 ≤ π‘˜, where 𝐻𝑗,0 = 𝐻𝑗 ,
(iii) the intervals 𝐻𝑗,𝑖 = 𝑓𝑖 (𝐻𝑗 ) are mutually disjoint for
1 ≤ 𝑖 ≤ 𝑛, 1 ≤ 𝑗 ≤ π‘˜,
(iv) π‘Ž < min(𝐻𝑗,𝑖 ) − 𝛾𝑗 πœ†(𝐻𝑗,𝑖 ) < max(𝐻𝑗,𝑖 ) + 𝛾𝑗 πœ†(𝐻𝑗,𝑖 ) < 𝑏,
for 0 ≤ 𝑖 ≤ 𝑛 − 1,
(v) max1≤𝑗≤π‘˜ {𝛾𝑗 , (𝑏 − π‘Ž)𝛾𝑗 } < (1/8)[min0≤𝑖≤𝑛, 1≤𝑗≤π‘˜ {πœ–/π‘˜,
πœ†(𝑓𝑖 (𝐻𝑗 )), 𝛽𝑗,𝑖 }], where 𝛽𝑗,𝑖 = inf{|𝑓󸀠 (π‘₯)| : π‘₯ ∈
𝑓𝑖 (𝐻𝑗 )}.
Let π‘₯0 = 𝑑𝑖 for some 𝑖, 𝛾, and 𝐽 = [𝑐, 𝑑] be the associated
𝛾𝑖 and 𝐻𝑖 , respectively. Define
𝛾
π‘₯ −𝑐
if π‘₯ = 𝑐 + 0
,
− ,
{
{
{ 3
2
{
{
{
{
{
3 (π‘₯0 − 𝑐)
{
{
0,
if π‘₯ = 𝑐 +
,
{
{
{
4
{
{
{
{
𝛾,
if π‘₯ = π‘₯0 ,
{
{
{
𝑑 − π‘₯0
𝑔𝐽,π‘₯0 ,𝛾 (π‘₯) = {0,
if π‘₯ = π‘₯0 +
,
{
4
{
{
{
{
{
𝛾
𝑑 − π‘₯0
{
{
{
if π‘₯ = π‘₯0 +
,
− ,
{
{ 3
2
{
{
{
{
{
{
0,
if π‘₯ ∈ [π‘Ž, 𝑏] \ (𝑐, 𝑑) ,
{
{
{linear, otherwise,
(13)
π‘₯ −𝑐
{
,
𝛾,
if π‘₯ = 𝑐 + 0
{
{
4
{
{
{
{
,
0,
if
π‘₯
=
π‘₯
{
0
{
{
{
3 (𝑑 − π‘₯0 )
π‘Ÿπ½,π‘₯0 ,𝛾 (π‘₯) = { −𝛾 (π‘₯0 − 𝑐)
{
if π‘₯ = π‘₯0 +
,
{
{
𝑑
−
π‘₯
4
{
0
{
{
{
0
if π‘₯ ∈ [π‘Ž, 𝑏] \ (𝑐, 𝑑) ,
{
{
{
otherwise,
{linear,
π‘₯
β„Ž1 (π‘₯) = ∫ 𝑔𝐽,π‘₯0 ,𝛾 (𝑑) 𝑑𝑑,
π‘Ž
π‘₯
β„Ž2 (π‘₯) = ∫ π‘Ÿπ½,π‘₯0 ,𝛾 (𝑑) 𝑑𝑑.
π‘Ž
σΈ€ 
σΈ€ 
We have β„Žπ‘š (𝑐) = β„Žπ‘š (𝑑) = β„Žπ‘š
(𝑐) = β„Žπ‘š
(𝑑) = 0 for π‘š = 1, 2,
σΈ€ 
β„Ž1 (π‘₯0 ) = 0, β„Ž1 (π‘₯0 ) = 𝛾, β„Ž2 (π‘₯0 ) = (π‘₯0 − 𝑐)𝛾/2, and β„Ž2σΈ€  (π‘₯0 ) = 0.
By considering four different cases, we construct a function
β„Ž ∈ 𝐢1 (𝐼, 𝐼) ∩ 𝐡(𝑓, πœ–) so that, for Cases A and B, 𝐹𝑛 (β„Ž) ∩ {π‘₯ :
(β„Žπ‘› )σΈ€  (π‘₯) = 1} ∩ 𝐽 = 0 and 𝐹𝑛 (β„Ž) ∩ {π‘₯ : (β„Žπ‘› )σΈ€  (π‘₯) = 1} ∩ 𝐽 =
{π‘₯0 } for Cases C and D. From conditions (iv) and (v) we have
β„Ž(𝐼) ⊂ 𝐼, and from condition (v) it follows that β„Ž is of the
same monotone type as 𝑓 on each 𝑓𝑠 (𝐽), 0 ≤ 𝑠 < 𝑛.
Case A. Let 𝑓𝑛 (π‘₯) < π‘₯ for 𝐽 \ {π‘₯0 } and 𝑓𝑛 (π‘₯0 ) = π‘₯0 .
(i) If 𝑓 is strictly increasing on 𝐽, then by taking
β„Ž(π‘₯) = 𝑓(π‘₯) − β„Ž2 (π‘₯) we have β„Ž(π‘₯) < 𝑓(π‘₯) on π½π‘œ
4
International Journal of Mathematics and Mathematical Sciences
and β„Ž(π‘₯) = 𝑓(π‘₯) on 𝐼 \ π½π‘œ and the function β„Žπ‘›−1 is
also strictly increasing on 𝑓(𝐽), and as a result β„Žπ‘› (π‘₯) <
𝑓𝑛 (π‘₯) ≤ π‘₯ on π½π‘œ . Thus 𝐹𝑛 (β„Ž) ∩ 𝐽 = 0.
(ii) If 𝑓 is strictly decreasing on 𝐽, then by taking β„Ž(π‘₯) =
𝑓(π‘₯) + β„Ž2 (π‘₯) we have β„Ž(π‘₯) > 𝑓(π‘₯) on π½π‘œ and β„Ž(π‘₯) =
𝑓(π‘₯) on 𝐼 \ π½π‘œ and the function β„Žπ‘›−1 is also strictly
decreasing on 𝑓(𝐽) and as a result, β„Žπ‘› (π‘₯) < 𝑓𝑛 (π‘₯) ≤ π‘₯
on π½π‘œ . Thus 𝐹𝑛 (β„Ž) ∩ 𝐽 = 0.
Case B. Let 𝑓𝑛 (π‘₯) > π‘₯ for 𝐽 \ {π‘₯0 } and 𝑓𝑛 (π‘₯0 ) = π‘₯0 .
If 𝑓 is strictly increasing on 𝐽, take β„Ž(π‘₯) = 𝑓(π‘₯) + β„Ž2 (π‘₯),
and if 𝑓 is strictly decreasing on 𝐽, take β„Ž(π‘₯) = 𝑓(π‘₯) − β„Ž2 (π‘₯).
Then similar to Case A, we may show that β„Žπ‘› (π‘₯) > 𝑓𝑛 (π‘₯) ≥ π‘₯
for π‘₯ ∈ π½π‘œ ; hence, 𝐹𝑛 (β„Ž) ∩ 𝐽 = 0.
Case C. Let 𝑓𝑛 (π‘₯) < π‘₯ for 𝑐 < π‘₯ < π‘₯0 , 𝑓𝑛 (π‘₯) > π‘₯ for π‘₯0 < π‘₯ <
𝑑, and 𝑓𝑛 (π‘₯0 ) = π‘₯0 .
(i) If 𝑓 is strictly increasing on 𝐽, then by taking β„Ž(π‘₯) =
𝑓(π‘₯) + β„Ž1 (π‘₯) we have β„Ž(π‘₯) < 𝑓(π‘₯) for 𝑐 < π‘₯ < π‘₯0
and β„Ž(π‘₯) > 𝑓(π‘₯) for π‘₯0 < π‘₯ < 𝑑, and the function
β„Žπ‘›−1 is also strictly increasing on 𝑓(𝐽). Thus β„Žπ‘› (π‘₯) <
𝑓𝑛 (π‘₯) < π‘₯ for 𝑐 < π‘₯ < π‘₯0 and β„Žπ‘› (π‘₯) > 𝑓𝑛 (π‘₯) > π‘₯ for
π‘₯0 < π‘₯ < 𝑑; hence, 𝐹𝑛 (β„Ž) ∩ 𝐽 = {π‘₯0 }.
(ii) If 𝑓 is strictly decreasing on 𝐽, then by taking β„Ž(π‘₯) =
𝑓(π‘₯) − β„Ž1 (π‘₯) we have β„Ž(π‘₯) > 𝑓(π‘₯) for 𝑐 < π‘₯ < π‘₯0
and β„Ž(π‘₯) < 𝑓(π‘₯) for π‘₯0 < π‘₯ < 𝑑, and the function
β„Žπ‘›−1 is also decreasing on 𝑓(𝐽). Thus β„Žπ‘› (π‘₯) < 𝑓𝑛 (π‘₯) <
π‘₯ for 𝑐 < π‘₯ < π‘₯0 and β„Žπ‘› (π‘₯) > 𝑓𝑛 (π‘₯) > π‘₯ for π‘₯0 < π‘₯ <
𝑑; hence, 𝐹𝑛 (β„Ž) ∩ 𝐽 = {π‘₯0 }.
Case D. Let 𝑓𝑛 (π‘₯) > π‘₯ for 𝑐 < π‘₯ < π‘₯0 , 𝑓𝑛 (π‘₯) < π‘₯ for π‘₯0 < π‘₯ <
𝑑, and 𝑓𝑛 (π‘₯0 ) = π‘₯0 .
If 𝑓 is strictly increasing on 𝐽, take β„Ž(π‘₯) = 𝑓(π‘₯) − β„Ž1 (π‘₯),
and if 𝑓 is strictly decreasing on 𝐽, take β„Ž(π‘₯) = 𝑓(π‘₯) + β„Ž1 (π‘₯).
Then similar to Case C, we may show that β„Žπ‘› (π‘₯) > 𝑓𝑛 (π‘₯) > π‘₯
for 𝑐 < π‘₯ < π‘₯0 and β„Žπ‘› (π‘₯) < 𝑓𝑛 (π‘₯) < π‘₯ for π‘₯0 < π‘₯ < 𝑑. Thus
𝐹𝑛 (β„Ž) ∩ 𝐽 = {π‘₯0 }.
Note that β„Žπ‘› is strictly increasing on 𝐽 ∪ 𝑓𝑛 (𝐽), so we have
𝐹𝑛 (β„Ž) ∩ [(𝐽 \ 𝑓𝑛 (𝐽)) ∪ (𝑓𝑛 (𝐽) \ 𝐽)] = 0.
(14)
Thus for π‘₯ ∈ 𝐼 either orb(π‘₯, β„Ž) = orb(π‘₯, 𝑓) or orb(π‘₯, β„Ž) ∩
𝐽 =ΜΈ 0, of which in such case 𝐹𝑛 (β„Ž)∩𝐽 = {π‘₯0 }, β„ŽσΈ€  (π‘₯0 ) = 𝑓󸀠 (π‘₯0 )±
𝛾, and β„ŽσΈ€  (β„Žπ‘– (π‘₯0 )) = 𝑓󸀠 (𝑓𝑖 (π‘₯0 )) for 1 ≤ 𝑖 ≤ 𝑛 − 1. Thus
σΈ€ 
(β„Žπ‘› ) (π‘₯0 )
= β„ŽσΈ€  (β„Žπ‘›−1 (π‘₯0 )) β„ŽσΈ€  (β„Žπ‘›−2 (π‘₯0 )) ⋅ ⋅ ⋅ β„ŽσΈ€  (β„Ž (π‘₯0 )) β„ŽσΈ€  (π‘₯0 )
= 𝑓󸀠 (𝑓𝑛−1 (π‘₯0 )) 𝑓󸀠 (𝑓𝑛−2 (π‘₯0 ))
× π‘“σΈ€  (𝑓𝑛−3 (π‘₯0 )) ⋅ ⋅ ⋅ 𝑓󸀠 (𝑓 (π‘₯0 )) β„ŽσΈ€  (π‘₯0 )
σΈ€ 
=
(𝑓𝑛 ) (π‘₯0 ) β„ŽσΈ€  (π‘₯0 ) 𝑓󸀠 (π‘₯0 ) ± 𝛾
=
𝑓󸀠 (π‘₯0 )
𝑓󸀠 (π‘₯0 )
= [1 ±
𝑓󸀠
𝛾
] =ΜΈ 1.
(π‘₯0 )
We have
σΈ€ 
𝐹𝑛 (β„Ž) ∩ {π‘₯ : (β„Žπ‘› ) (π‘₯) = 1}
σΈ€ 
= [𝐹𝑛 (𝑓) ∩ {π‘₯ : (𝑓𝑛 ) (π‘₯) = 1}] \ {π‘₯0 } .
(16)
Also
πœ–
󡄨
󡄨
𝜌0 (𝑓󸀠 , β„ŽσΈ€  ) ≤ sup {󡄨󡄨󡄨𝑔 (π‘₯)󡄨󡄨󡄨 , |π‘Ÿ (π‘₯)|} ≤
,
8π‘˜
π‘₯∈𝐽
πœ–
󡄨 󡄨
󡄨
󡄨
𝜌0 (𝑓, β„Ž) ≤ sup {σ΅„¨σ΅„¨σ΅„¨β„Ž1 (π‘₯)󡄨󡄨󡄨 , σ΅„¨σ΅„¨σ΅„¨β„Ž2 (π‘₯)󡄨󡄨󡄨} ≤
.
8π‘˜
π‘₯∈𝐽
(17)
Hence 𝜌1 (𝑓, β„Ž) ≤ 𝜌0 (𝑓, β„Ž) + 𝜌0 (𝑓󸀠 , β„ŽσΈ€  ) ≤ πœ–/4π‘˜.
Doing this recursively for each 𝑑𝑗 , 1 ≤ 𝑗 ≤ π‘˜, we get a
function 𝑔 ∈ 𝐡(𝑓, πœ–/4) such that 𝐹𝑛 (𝑔) ∩ {π‘₯ : (𝑔𝑛 )σΈ€  (π‘₯) = 1} =
0.
Theorem 8. Typically continuously differentiable self-maps of
intervals have a countable set of periodic points.
Proof. Take
𝐻𝑛 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : 𝐹𝑛 (𝑓) is not finite} ,
σΈ€ 
𝐡1 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : π‘Ž ∈ 𝑃𝑓2 ∩ {π‘₯ : (𝑓2 ) (π‘₯) = 1}
with 𝑓 (π‘Ž) = 𝑏} ,
𝐹1 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : ∃π‘₯0 ∈ 𝐼 such that
π‘₯0 ∈ 𝐹1 (𝑓) and 𝑓󸀠 (π‘₯0 ) = 1} ,
σΈ€ 
E𝑛 = {𝑓 ∈ 𝐢1 (𝐼, 𝐼) : 𝐹𝑛 (𝑓) ∩ {π‘₯ : (𝑓𝑛 ) (π‘₯) = 1} =ΜΈ 0} .
(18)
The sets 𝐹1 and 𝐡1 are first category sets (see Theorems
5 and 6). For 𝑛 ≥ 1 from Lemmas 4 and 1 we have that E𝑛
is closed and 𝐻𝑛 ⊆ E𝑛 . Without loss of generality we may
assume that 𝑓 ∈ 𝐢1 (𝐼, 𝐼) is neither a constant function nor
a polynomial of first degree, and then by Lemma 2 we may
choose a polynomial 𝑔 ∈ 𝐢1 (𝐼, 𝐼) of degree 𝑛 ≥ 2 such
that 𝑔(𝐼) ⊂ πΌπ‘œ and 𝜌1 (𝑓, 𝑔) < πœ–/4. Using Theorem 7 we can
construct β„Ž ∈ 𝐢1 (𝐼, 𝐼) so that 𝜌1 (𝑔, β„Ž) < πœ–/4 and the set
𝐴 𝑛 = 𝐹𝑛 (β„Ž) ∩ {π‘₯ : (β„Žπ‘› )σΈ€  (π‘₯) = 1} = 0, thus 𝜌1 (𝑓, β„Ž) < πœ–
and β„Ž ∉ E𝑛 , hence Eπ‘œπ‘› = 0. This implies that for each π‘˜ ≥ 2
the set Eπ‘˜ is nowhere dense. Thus 𝐹 = (∪∞
π‘˜=2 Eπ‘˜ ) ∪ 𝐹1 ∪ 𝐡1 is
1
a first category set, so 𝑀 = 𝐢 (𝐼, 𝐼) \ 𝐹 is a residual set, and,
for 𝑓 ∈ 𝑀, 𝑃(𝑓) = ∪∞
𝑛=1 𝐹𝑛 (𝑓) is countable.
Acknowledgments
(15)
The author would like to thank professor David Preiss for his
valuable comments and suggestions as well as the anonymous
referees for their comments.
International Journal of Mathematics and Mathematical Sciences
References
[1] S. J. Agronsky, A. M. Bruckner, and M. Laczkovich, “Dynamics
of typical continuous functions,” Journal of the London Mathematical Society, vol. 40, no. 2, pp. 227–243, 1989.
[2] A. A. Alikhani-Koopaei, “On the size of the sets of chain
recurrent points, periodic points, and fixed points of functions,”
Global Journal of Pure and Applied Mathematics, vol. 4, no. 2, pp.
113–121, 2008.
[3] A. A. Alikhani-Koopaei, “On periodic and recurrent points
of continuous functions,” in Proceedings of the 28th Summer
Symposium Conference, Real Analysis Exchange, pp. 37–40,
2004.
[4] A. Alikhani-Koopaei, “On common fixed points, periodic
points, and recurrent points of continuous functions,” International Journal of Mathematics and Mathematical Sciences, vol.
2003, no. 39, pp. 2465–2473, 2003.
[5] T. H. Steele, “A note on periodic points and commuting
functions,” Real Analysis Exchange, vol. 24, no. 2, pp. 781–789,
1998-1999.
[6] A. Alikhani-Koopaei, “On common fixed and periodic points
of commuting functions,” International Journal of Mathematics
and Mathematical Sciences, vol. 21, no. 2, pp. 269–276, 1998.
[7] A. J. Schwartz, “Common periodic points of commuting functions,” The Michigan Mathematical Journal, vol. 12, pp. 353–355,
1965.
5
Advances in
Operations Research
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Advances in
Decision Sciences
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Mathematical Problems
in Engineering
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Journal of
Algebra
Hindawi Publishing Corporation
http://www.hindawi.com
Probability and Statistics
Volume 2014
The Scientific
World Journal
Hindawi Publishing Corporation
http://www.hindawi.com
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
International Journal of
Differential Equations
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Volume 2014
Submit your manuscripts at
http://www.hindawi.com
International Journal of
Advances in
Combinatorics
Hindawi Publishing Corporation
http://www.hindawi.com
Mathematical Physics
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Journal of
Complex Analysis
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
International
Journal of
Mathematics and
Mathematical
Sciences
Journal of
Hindawi Publishing Corporation
http://www.hindawi.com
Stochastic Analysis
Abstract and
Applied Analysis
Hindawi Publishing Corporation
http://www.hindawi.com
Hindawi Publishing Corporation
http://www.hindawi.com
International Journal of
Mathematics
Volume 2014
Volume 2014
Discrete Dynamics in
Nature and Society
Volume 2014
Volume 2014
Journal of
Journal of
Discrete Mathematics
Journal of
Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com
Applied Mathematics
Journal of
Function Spaces
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Optimization
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Hindawi Publishing Corporation
http://www.hindawi.com
Volume 2014
Download