S Convexity Phenomenon III and Insights On

Notes on one of the generators of the
S−convexity phenomenon III and insights on

the application of our previous geometric
results
I.M.R. Pinheiro
PO BOX 12396, A’Beckett st, Melbourne, Victoria, 8006, AUSTRALIA
e-mail: mrpprof essional@yahoo.com
Abstract: In this one more paper, we try to create new models for the S-
convexity phenomenon. As a side result, we nullify three previous results
connected to the concept.

AMS Subj. Classification:26D10(Primary), 26D15 (Secondary).
Key-words: Generator, S-convexity, Examples, Dragomir, Pierce.
1. Introduction
So far, we seem to be stuck with the polynomial model, when it comes to find
examples of S−convex functions. However, in this paper, we will dare putting
down this paradigm and provide the reader with trigonometric models as well.
In the sections that follow, we deal with:
• Terminology;
• Definitions;
• Simple and basic trigonometric examples, which may be combined;

• Review of past results;
• Conclusion.
2. Terminology
We use the same symbols and definitions presented in Pinheiro [2]:
• Ks1 for the class of S-convex functions in the first sense, some S;
• Ks2 for the class of S-convex functions in the second sense, some S;
• K0 for the class of convex functions;
• s1 for the variable S, 0 < S ≤ 1, used in the first definition of S-

convexity;
• s2 for the variable S, 0 < S ≤ 1, used in the second definition of S-

convexity.
Remark 1. The class of 1-convex functions is just a restriction of the class
of convex functions, that is, when X = <+ ,
K11 ≡ K12 ≡ K0 .
3. Definitions
Definition 3. A function f : X− > <, f ∈ C 1 , is said to be s1 -convex if

the inequality
1
f (λx + (1 − λs ) s y) ≤ λs f (x) + (1 − λs )f (y)
holds ∀λ ∈ [0, 1], ∀x, y ∈ X such that X ⊂ <+ .
Definition 4. f is called s2 -convex, s 6= 1, if the graph lies below a ‘bent

chord’ (L) between any two points, that is, for every compact interval J ⊂ I,
with boundary ∂J, it is true that
sup(L − f ) ≥ sup(L − f ).
J ∂J
Definition 5. A function f : X− > <, in C 1 , is said to be s2 -convex if
the inequality
f (λx + (1 − λ)y) ≤ λs f (x) + (1 − λ)s f (y)
holds ∀λ ∈ [0, 1], ∀x, y ∈ X such that X ⊂ <+ .

4. Simple new model, trigonometric
From past work, we will use one conclusion, one definition (geometric), and a
theorem (from [3], [2], and [4]):
Conclusion:
21−s is always the maximum height for our curve.
Theorem 5.1. The sum of two S−convex functions is also an S−convex
function.
Lemma 1. If a function is represented graphically by a smooth curve,
continuous, and the distance between the straight line formed between the couple
of points in the curve, which form the highest pitch for it, and the actual
function curve is at most one, as seen in [3], we then have an S−convex curve.
Proof. Such is obvious and follows directly from the definition combined with
our explanations and deductions proven in [3].
We now consider a few trigonometric functions on the grounds of their suit-
ability for being mathematical models of S−convex functions.
• sin(x) and cos(x).
Notice that it is redundant analyzing both functions here. Their graph-

ical behavior, which is the only thing that matters for S−convexity, is
practically the same, only different regarding time of occurrence. There-
fore, whatever we deduce regarding one of them may be applied directly

to another, with subtle adaptation to domain intervals (ninety degrees).
The highest pitch of the sine function is achieved at one. With this, we
already know that it is not possible to consider the first quadrant as a
whole for our example, once 21−s will never reach the value two, only
go close (given the limitations imposed upon the value of s). However,
suffices taking a slightly shorter interval in the domain and the example
is achieved. Say we then pick [0, Π2 ), once that will give us the precise
maximum fit. What really matters is the most bent piece of the sine
curve between those points of the domain. In this case, half of it. Our
distance then, regarding the possible straight line between those points
and the origin will be 1 (app. result for distance between origin and
the midpoint in the domain, in what regards its image) and the distance
between the actual curve and the origin will be 1.26 (app.). Therefore,
we hold 0.26 of distance, to be put against our 21−s , to find out our most
suitable value for s. In this case, we end up with 0.3 with some other
decimal figures after the 3. Therefore, we may state that it belongs to
all s values after, and smaller, than 0.3, for instance, with no mistake.
Theorem 6. cos(x) is a model for Ks2 , if the domain is restricted to

[0, Π2 ), or (exclusively) ( Π2 , Π], or (exclusively) [ Π2 , 3Π
2
] (when it is con-
vex), or (exclusively) ( 3Π
2
, 2Π], as well as all other determinations (not
minimal) of the mentioned intervals of domain.
Proof. As above.
Corollary 1. sin(x) is a model for Ks2 , if the domain is restricted to
(0, Π2 ], or (exclusively) [ Π2 , Π), or (exclusively) (Π, 3Π
2
], or (exclusively)
[ 3Π
2
, 2Π).
5. Review of past works
FFFFNullification of three past resultsFFFF
From [5] we read, on page 687:
Theorem 6.1. Let 0 < s < 1. If f ∈ Ks2 then f is nonnegative.
There is a major mistake in understanding regarding the own definition of
both convexity, and its proper geometric extension, Ks2 , which clearly appears
in the proof. Dragomir et al. makes use of the same point of the domain
in order to compare the chord formed between two points of the domain and
the actual curve of the function (because they are geometric definitions, well
matched with the analytical ones, not only analytical). Two points, in order
to form a chord, cannot, possibly, mean one...
Therefore, such result does not hold at all. Obviously the case that if it ever
held for Ks2 , it would also have to hold for K12 , or K0 , as defined before.
From [5] we read, on page 688:
Theorem 6.2. Let f be a function on [a, b] which is s-convex in the second
sense. Then for a < y < z < b we have
|f (y) − f (z)| ≤ (z − y)s max(f (b)/(b − y)s , f (a)/(z − a)s )
so the f is locally Hölder continuous of orders s on (a, b). Thus f is Riemann

integrable on [a, b].
There is a major problem, once more, with the proof of such theorem, as
appears in [5]. Basically, one cannot first create a new variable, call it t, with
the ‘attrib’ command from Maple, and define it based on other three already
existing variables, as for the own theorem start, then make use of one of the
variables, in which the creation of t was based, in the same expression as t, in-
stead of expressing z as depending on same variables as rest of the expression.

However, such step could represent minor typos, or distractions, and the proof
still be accurate. So, first of all, re-writing the first inequality which appears
on that proof, we get: f ((1 − t)y + tb) ≤ (1 − t)s f (y) + ts f (b) ≤ f (y) + ts f (b).
Given we impose Ks2 pertinence to f , such would be correct, in principle. Next
step, re-written, would be: f ((1 − t)y + tb) − f (y) ≤ ts f (b), which still makes
sense. However, if ever willing to call whatever comes in brackets z, one would
also have to change y into its expression depending on z, for that is for the
best of Mathematics. On top of that, b would also appear expressed in terms
of z. Such will actually invalidate the conclusion there present.
It is obviously worthless continuing on that line of thought. However, there is a
major absurdity, not to say insanity, in the mentioned theorem (actually more
than one): Hölder continuity refers to a constant multiplied by the difference
between the variables mentioned to the left of the inequality...not to a function
of the variables instead...Please, if in doubt, check sources such as [1]...
It also does not make sense, even if that step was not there (no constants to
right), writing that that implies Riemann integrability in a theorem, this is
simply inadequate.
So, basically, that was never a theorem and should simply be erased from
mathematical history, in full, as if it had never been published, refereed by so
many people, from a whole editorial board...for the sake of ethics, we should
simply forget it.
Still from [5] we read, on page 688:
Theorem 6.3. Let f be a function on [a, b] which is s-convex in the second
sense. If f (c) = 0 for some c ∈ [a, b] then f (x) ≤ f (y) if c ≤ x ≤ y ≤ b and

f (x) ≥ f (y) if a ≤ x ≤ y ≤ c.
Well, from our past work and deductions, the theorem above is absurd.
However, we also look into the details of the proof, not to say we do not hold
consideration and respect for someone else’s published work. Science must be
about argumentation and who is technically sound.
So, there we go: The first line of the proof states that ‘If c ≤ x ≤ y ≤ b then
f (x) ≤ ts f (c) + (1 − t)s f (y) if x = tc + (1 − t)y. We may stop here, for the sake
of all of us only reading the same remark twice. Once more, x, the variable,
is defined based on other two, and, in the definition of x, c, t, and y are all
included. However, in the inequality formed, x appears as if it is independent

from whatever appears to right side...
If x were adequately mentioned in the inequality, however, the conclusion

would not follow, only something else. It is obvious that the last inequality
attained in the ‘proof’ would bring x in both sides, if Mathematics were ever
considered...
6. Conclusion
In this paper, we present a new model for S−convex functions and, by pre-
senting it, we believe to have exhausted the issues regarding finding examples
for functions in Ks2 for any interested researcher, once it will suffice repeat our
procedures here, with the graph of functions, to determine, same way we do
with convexity, whether a certain function belongs, or not, to Ks2 (only). On
the top of that, we review and nullify three previous results by Dragomir et
al. ([5]).
7. Bibliography
[1] J. B. Brown; G. Kozlowski, Smooth Interpolation, Holder Continuity, and

the Takagi-van der Waerden Function. The American Mathematical Monthly,
Vol. 110, No. 2 (Feb., 2003), pp. 142-147.
[2] M. R. Pinheiro, Exploring the Concept of S-convexity. Aequationes Math-
ematicae. 74, 3: 201-209 (2007).
[3] M. R. Pinheiro, S-convexity as a local approximation to convex functions.

Submitted, 2005. Online preprint at www.geocities.com/mrpprof essional.
[4] M. R. Pinheiro, Notes on one of the generators of the S−convexity phe-

nomenon III. Submitted, 2008. Online preprint at www.geocities.com/msorf iap2.
[5] S. S. Dragomir; S. Fitzpatrick The Hadamard Inequalities for S-convex
Functions in the Second Sense. Demonstratio Mathematica, V. XXXII, Nr. 4,

1999.

S Convexity Phenomenon III and Insights On

Uploaded by

Copyright:

Available Formats

S Convexity Phenomenon III and Insights On

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S Convexity Phenomenon III and Insights On

Uploaded by

Copyright:

Available Formats

Notes on one of the generators of the

S−convexity phenomenon III and insights on

PO BOX 12396, A’Beckett st, Melbourne, Victoria, 8006, AUSTRALIA

e-mail: mrpprof essional@yahoo.com

convexity phenomenon. As a side result, we nullify three previous results

connected to the concept.

In the sections that follow, we deal with:

• Simple and basic trigonometric examples, which may be combined;

We use the same symbols and definitions presented in Pinheiro [2]:

• K0 for the class of convex functions;

• s1 for the variable S, 0 < S ≤ 1, used in the first definition of S-

• s2 for the variable S, 0 < S ≤ 1, used in the second definition of S-

Remark 1. The class of 1-convex functions is just a restriction of the class

of convex functions, that is, when X = <+ ,

Definition 3. A function f : X− > <, f ∈ C 1 , is said to be s1 -convex if

holds ∀λ ∈ [0, 1], ∀x, y ∈ X such that X ⊂ <+ .

Definition 4. f is called s2 -convex, s 6= 1, if the graph lies below a ‘bent

Definition 5. A function f : X− > <, in C 1 , is said to be s2 -convex if

f (λx + (1 − λ)y) ≤ λs f (x) + (1 − λ)s f (y)

holds ∀λ ∈ [0, 1], ∀x, y ∈ X such that X ⊂ <+ .

21−s is always the maximum height for our curve.

Theorem 5.1. The sum of two S−convex functions is also an S−convex

Lemma 1. If a function is represented graphically by a smooth curve,

We now consider a few trigonometric functions on the grounds of their suit-

ability for being mathematical models of S−convex functions.

• sin(x) and cos(x).

Notice that it is redundant analyzing both functions here. Their graph-

practically the same, only different regarding time of occurrence. There-

fore, whatever we deduce regarding one of them may be applied directly

Theorem 6. cos(x) is a model for Ks2 , if the domain is restricted to

5. Review of past works

FFFFNullification of three past resultsFFFF

From [5] we read, on page 687:

Theorem 6.1. Let 0 < s < 1. If f ∈ Ks2 then f is nonnegative.

There is a major mistake in understanding regarding the own definition of

From [5] we read, on page 688:

Theorem 6.2. Let f be a function on [a, b] which is s-convex in the second

sense. Then for a < y < z < b we have

|f (y) − f (z)| ≤ (z − y)s max(f (b)/(b − y)s , f (a)/(z − a)s )

so the f is locally Hölder continuous of orders s on (a, b). Thus f is Riemann

stead of expressing z as depending on same variables as rest of the expression.

of the variables instead...Please, if in doubt, check sources such as [1]...

right), writing that that implies Riemann integrability in a theorem, this is

mathematical history, in full, as if it had never been published, refereed by so

Still from [5] we read, on page 688:

Theorem 6.3. Let f be a function on [a, b] which is s-convex in the second

sense. If f (c) = 0 for some c ∈ [a, b] then f (x) ≤ f (y) if c ≤ x ≤ y ≤ b and

included. However, in the inequality formed, x appears as if it is independent

If x were adequately mentioned in the inequality, however, the conclusion

[1] J. B. Brown; G. Kozlowski, Smooth Interpolation, Holder Continuity, and

Vol. 110, No. 2 (Feb., 2003), pp. 142-147.

[2] M. R. Pinheiro, Exploring the Concept of S-convexity. Aequationes Math-

ematicae. 74, 3: 201-209 (2007).

[3] M. R. Pinheiro, S-convexity as a local approximation to convex functions.

[4] M. R. Pinheiro, Notes on one of the generators of the S−convexity phe-

[5] S. S. Dragomir; S. Fitzpatrick The Hadamard Inequalities for S-convex

Functions in the Second Sense. Demonstratio Mathematica, V. XXXII, Nr. 4,

You might also like