Next Article in Journal
Identification of Couplings in Adaptive Dynamical Networks of Time-Delayed Feedback Oscillators
Previous Article in Journal
Global Dynamics of an SEIR Model with the Age of Infection and Vaccination
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Asymptotic Stability of the Pexider–Cauchy Functional Equation in Non-Archimedean Spaces

1
Department of Mathematics, Faculty of Sciences, University of Mohaghegh Ardabili, Ardabil 56199-11367, Iran
2
Humanitas College, Kyung Hee University, Yongin 17104, Korea
*
Author to whom correspondence should be addressed.
These authors contributed equally to this work.
Mathematics 2021, 9(18), 2197; https://doi.org/10.3390/math9182197
Submission received: 24 August 2021 / Revised: 3 September 2021 / Accepted: 7 September 2021 / Published: 8 September 2021
(This article belongs to the Section Difference and Differential Equations)

Abstract

:
In this paper, we investigated the asymptotic stability behaviour of the Pexider–Cauchy functional equation in non-Archimedean spaces. We also showed that, under some conditions, if f ( x + y ) g ( x ) h ( y ) ε , then f , g and h can be approximated by additive mapping in non-Archimedean normed spaces. Finally, we deal with a functional inequality and its asymptotic behaviour.

1. Introduction

The classical problem of the stability of homomorphisms was first posed by Ulam [1] in 1940 as follows: Given a group ( G 1 , ) , a metric group ( G 2 , , d ) and ε > 0 . Does δ > 0 exist such that if the function f : G 1 G 2 satisfies the inequality:
d ( f ( x y ) , f ( x ) f ( y ) ) < δ , x , y G 1 ,
then there exists a homomorphism T : G 1 G 2 such that d ( f ( x ) , T ( x ) ) < ε for all x G 1 ? In 1941, Donald H. Hyers [2] gave a partial affirmative answer to this question under the assumption that f : X Y is a function between Banach spaces X and Y. Indeed, D. H. Hyers’ classical theorem states that, if X is a linear space and Y is a Banach space and, for some ε 0 , a function f : X Y is ε -additive, meaning that:
f ( x + y ) f ( x ) f ( y ) ε , x , y X
then there exists a unique additive function A : X Y such that f is ε -close to A, i.e., f ( x ) A ( x ) ε for all x X . Ulam’s question can be expressed in a more general way as follows: when is it true that a function which approximately satisfies a functional equation ( ) must be close to an exact solution of ( ) ?
Aoki [3] and Th. M. Rassias [4] extended the Hyers’ theorem by considering an unbounded Cauchy difference. They tried to weaken the condition for the bound of the norm of the Cauchy difference in (1) as follows:
f ( x + y ) f ( x ) f ( y ) ε ( x p + y p ) , x , y X
where 0 p < 1 . In 1994, Găvruta [5] provided a generalization of Rassias’ theorem by replacing the bound ε ( x p + y p ) by a general control function φ ( x , y ) . In recent decades, several stability problems for various functional equations and also for mappings with more general domains and ranges have been investigated by a number of mathematicians. We refer the interested reader the following books and surveys [6,7,8,9,10] and the references therein for more detailed information.
Let us recall (see, for instance, [11]) some basic definitions and facts concerning non-Archimedean normed spaces. By a non-Archimedean field, we mean a field F equipped with a function (valuation) | . | from F into [ 0 , + ) such that:
( i )
| r | = 0 if and only if r = 0 ;
( i i )
| r s | = | r | | s | , r , s F ;
( i i i )
| r + s | max { | r | , | s | } , r , s F .
Clearly | 1 | = | 1 | = 1 and | n | 1 for all integers n. An obvious example is the trivial valuation that can be defined on any field F by setting:
| x | : = 0 , if x = 0 ; 1 , if x 0 .
Example 1.
For a given prime number p, any non-zero rational number x can be uniquely written as x = a b p n x , where a , b , n x Z and a , b are integers not divisible by p. Then, the function | . | p : Q [ 0 , + ) defined by
| x | : = 0 , if x = 0 ; p n x , if x 0 .
is a non-Archimedean (nontrivial) valuation on ℚ.
Let X be a vector space over a non-Archimedean field F with a non-trivial valuation | . | . A function . : X [ 0 , + ) is called a non-Archimedean norm (valuation) if it satisfies the following conditions:
( i )
x = 0 if and only if x = 0 ;
( i i )
r x = | r | x for all r F and all x X ;
( i i i )
the strong triangle inequality; namely,
x + y max { x , y } , x , y X .
Then, ( X , . ) is called a non-Archimedean normed space. A sequence { x n } n = 1 in X is called a Cauchy sequence if lim m , n x n x m = 0 . A sequence { x n } n = 1 in X is said to converge to x X if lim n x n x = 0 . By the strong triangle inequality ( i i i ) , we have
x n x m max x j + 1 x j : m j n 1 , n > m .
Then, the sequence { x n } n = 1 in a non-Archimedean space is Cauchy if and only if { x n + 1 x n } n = 1 converges to zero. By a Banach non-Archimedean space, we mean that every Cauchy sequence is convergent. It is necessary to remember that | . | p is a non-Archimedean norm on ℚ, and that the completion of ℚ with respect to | . | p is denoted by Q p and is called the p-adic number field. Indeed, p-adic numbers do not satisfy the Archimedean axiom which states: for each x , y > 0 , there exists an integer n such that n x > y .
In [12], the authors showed that if f : Q p R is a continuous ε -additive function, then there exists a unique additive function A : X Y such that f is ε -close to A. In [13,14], the authors investigated the stability of some functional equations in non-Archimedean normed spaces. In this paper, we investigated the asymptotic stability behaviour of the Pexider–Cauchy functional equation in non-Archimedean spaces. We will use the methods and ideas presented in [15,16,17].

2. Main Results

Throughout this section, we assume that X and Y are the non-Archimedean normed spaces over a non-Archimedean field F with a valuation | . | .
Theorem 1.
Let ε 0 and f , g , h : X Y satisfy the following asymptotic stability property:
lim sup min { x , y } f ( x + y ) g ( x ) h ( y ) ε .
If | n | < 1 for some integer n, then:
f ( x + y ) f ( x ) f ( y ) + f ( 0 ) ε , x , y X .
Proof. 
Let η > ε be an arbitrary real number. Then, by (2) there exists d > 0 such that:
f ( x + y ) g ( x ) h ( y ) < η , x , y d .
Let x , y X be arbitrary elements. Since | n | < 1 for some integer n, we obtain:
lim m + 1 n m z = +
for some z X   \   { 0 } . Hence we can choose u X such that u d + x + y . It is easy to see that:
min x u , y + u , u d .
Applying (4) and the strong triangle inequality, we obtain:
f ( x + y ) f ( x ) f ( y ) + f ( 0 ) max { g ( x u ) + h ( u ) f ( x ) , g ( u ) + h ( y + u ) f ( y ) , f ( x + y ) g ( x u ) h ( y + u ) , f ( 0 ) g ( u ) h ( u ) } < η
for all x , y X . Taking the limit as η ε in the last inequality, we obtain (3): □
Corollary 1.
Let ε 0 and f : X Y satisfy the following asymptotic stability property:
lim sup min { x , y } f ( x + y ) f ( x ) f ( y ) ε .
If | n | < 1 for some integer n, then:
f ( x + y ) f ( x ) f ( y ) ε , x , y X .
Proof. 
Let η > ε be an arbitrary real number. By (5), there exists: d > 0 such that
f ( x + y ) f ( x ) f ( y ) < η , x , y d .
By the proof of Theorem 1, we obtain:
f ( x + y ) f ( x ) f ( y ) + f ( 0 ) < η , x , y X .
By using the strong triangle inequality, it follows from (7) and (8) that f ( 0 ) < η . Then:
f ( x + y ) f ( x ) f ( y ) max f ( x + y ) f ( x ) f ( y ) + f ( 0 ) , f ( 0 ) < η
for all x , y X . Taking the limit as η ε in the last inequality, we obtain (6): □
Corollary 2.
Let | n | < 1 for some integer n, and f , g , h : X Y fulfil the following asymptotic property:
lim min { x , y } f ( x + y ) g ( x ) h ( y ) = 0 .
Then, f is affine. Moreover, if f = g = h , then f is additive.
The following example shows that with the conditions of Corollary 2, functions g and h may not be affine even if g = h .
Example 2.
Let ℚ be the field of rational numbers and let | . | p be the p-adic absolute value, where p > 2 is a prime number. We define f , g , h : Q Q by
f ( x ) = x and g ( x ) = h ( x ) = x + 1 1 + x 2 , x Q .
Let n > 1 be an integer. For | x | p , | y | p p n , we have:
| f ( x + y ) g ( x ) h ( y ) | p = 1 1 + x 2 + 1 1 + y 2 p max 1 | 1 + x 2 | p , 1 | 1 + y 2 | p 1 p 2 n 1 0 , as n .
Then:
lim min { | x | p , | y | p } | f ( x + y ) g ( x ) h ( y ) | p = 0 ,
but g and h are not affine.
Example 3.
Let p > 2 be a prime number. Suppose that a function f : Q Q satisfies
f ( 1 ) = 1 and lim min { | x | p , | y | p } | f ( x + y ) f ( x ) f ( y ) | p = 0 .
Then f is the identity, i.e., f ( x ) = x for all x X .
Theorem 2.
Let ε 0 and Y be a Banach non-Archimedean space. Suppose that f , g , h : X Y are functions that satisfy:
lim sup min { x , y } f ( x + y ) g ( x ) h ( y ) ε .
If | n | < 1 for some integer n, then there exists a unique additive mapping A : X Y such that:
f ( x ) f ( 0 ) A ( x ) ε , x X .
Proof. 
It follows by Theorem 1 that:
f ( x + y ) f ( x ) f ( y ) + f ( 0 ) ε , x , y X .
Defining F : X Y by F ( x ) : = f ( x ) f ( 0 ) , we obtain:
F ( x + y ) F ( x ) F ( y ) ε , x , y X .
By ([13] [Theorem 2.1]), there is a unique additive mapping A : X Y such that F ( x ) A ( x ) ε for all x X (see also [14,18]). This completes the proof. □
By Corollary 1 and using the argument in Theorem 2, we obtain the following result.
Corollary 3.
Let ε 0 and Y be a Banach non-Archimedean space. Let f : X Y be functions satisfying:
lim sup min { x , y } f ( x + y ) f ( x ) f ( y ) ε .
If | n | < 1 for some integer n, then there exists a unique additive mapping A : X Y such that:
f ( x ) A ( x ) ε , x X .
Theorem 3.
Let ε 0 and f , g , h : X Y satisfy the following asymptotic stability property:
lim sup x + y f ( x + y ) g ( x ) h ( y ) ε .
If | n | < 1 for some integer n, then:
f ( x + y ) g ( x ) h ( y ) ε , x , y X .
Proof. 
Let η > ε be an arbitrary real number. Then, by (14) there exists: d > 0 such that
f ( x + y ) g ( x ) h ( y ) < η , x + y d .
Let x , y X be arbitrary elements. Since | n | < 1 for some integer n, we can choose u X such that 2 u d + x + y . It is easy to verify that:
x u + y + u d ; 2 u + x u d ; y + 2 u d ; y + u + u d ; x + u d .
Then, (16) and (17) yield:
f ( x + y ) g ( x ) h ( y ) max { f ( x + y ) g ( y + u ) h ( x u ) , f ( x + u ) g ( 2 u ) h ( x u ) , f ( y + 2 u ) g ( 2 u ) h ( y ) , f ( y + 2 u ) g ( y + u ) h ( u ) , f ( x + u ) g ( x ) h ( u ) } < η
for all x , y X . Taking the limit as η ε in the last inequality, we obtain (15). □
Corollary 4.
Let ε 0 and f , g , h : X Y satisfy the following asymptotic property:
lim x + y f ( x + y ) g ( x ) h ( y ) = 0
If | n | < 1 for some integer n, then f , g and h are affine. Moreover, if g ( 0 ) = h ( 0 ) = 0 , then f = g = h are additive.
Theorem 4.
Let ε 0 and f , g , h : X Y be functions satisfying (14). If | n | < 1 for some integer n, then there exists a unique additive mapping A : X Y such that:
f ( x ) f ( 0 ) A ( x ) ε , g ( x ) g ( 0 ) A ( x ) ε , h ( x ) h ( 0 ) A ( x ) ε
for all x X .
Proof. 
By Theorem 3, we have f ( x + y ) g ( x ) h ( y ) ε for all x , y X . Then, the result easily follows from ([13] [Theorem 2.4]) (see also [18]). □

3. Results on the Hyperstability

Theorem 5.
Let φ : [ 0 , + ) R be a function such that lim t + φ ( t ) = + and let f , g , h : X Y satisfy:
lim sup x + y φ ( x y ) f ( x + y ) g ( x ) h ( y ) < .
If | n | < 1 for some integer n, then f ( x + y ) = g ( x ) + h ( y ) for all x , y X , and consequently f , g and h are affine. Furthermore, f f ( 0 ) = g g ( 0 ) = h h ( 0 ) .
Proof. 
According to (19), there exist constants d > 0 and K > 0 such that:
φ ( x y ) f ( x + y ) g ( x ) h ( y ) < K , x + y d .
Let ε > 0 be arbitrary. We can choose N > 0 such that φ ( t ) K ε for all t N . Then, (20) implies that:
f ( x + y ) g ( x ) h ( y ) < ε , x + y d and x y N .
Let x , y X with x + y 2 d + 2 N . Since | n | < 1 for some integer n, we can choose u X such that:
min { 2 u , 3 u } d + N + x + y + x y + x 2 y .
It is easy to verify that:
min { y ± u , x ± u , x y + 2 u , u } d , min { y x ± 2 u , 3 u x , 2 u y , x u , x 2 y + 2 u , x 2 y + 3 u } N .
Since x + y 2 d + 2 N , two cases arise according to whether x d + N or y d + N . First, take the case x d + N . Then, applying (21), we obtain:
f ( x + y ) g ( x ) h ( y ) max { f ( x + y ) g ( x + u ) h ( y u ) , f ( x + 2 u ) g ( x + u ) h ( u ) , f ( x + u ) g ( x y + 2 u ) h ( y u ) , f ( x + u ) g ( x ) h ( u ) , f ( x + 2 u ) g ( x y + 2 u ) h ( y ) } < ε .
Now, consider the case y d + N . Then, applying (21) yields:
f ( x + y ) g ( x ) h ( y ) max { f ( x + y ) g ( y + u ) h ( x u ) , f ( x + u ) g ( 2 u ) h ( x u ) , f ( y + 2 u ) g ( 2 u ) h ( y ) , f ( y + 2 u ) g ( y + u ) h ( u ) , f ( x + u ) g ( x ) h ( u ) } < ε .
Therefore, f ( x + y ) g ( x ) h ( y ) ε for x + y 2 d + 2 N . Then, by Theorem 3, we have f ( x + y ) g ( x ) h ( y ) ε for all x , y X . Since ε > 0 was arbitrary, this inequality implies that f ( x + y ) = g ( x ) + h ( y ) for all x , y X . □
Remark 1.
Using the argument presented above, it can be concluded that Theorem 5 still holds if we have:
lim sup x + y φ max { x , y } f ( x + y ) g ( x ) h ( y ) <
instead of (20).
Theorem 6.
Let φ : [ 0 , + ) R be a function such that lim t + φ ( t ) = + and let f , g , h : X Y satisfy:
lim sup min { x , y } φ ( x y ) f ( x + y ) g ( x ) h ( y ) < .
If | n | < 1 for some non-zero integer n, then f is affine.
Proof. 
According to (22), there exist constants d > 0 and K > 0 such that:
φ ( x y ) f ( x + y ) g ( x ) h ( y ) < K , x , y d .
Let ε > 0 be arbitrary. We can choose N > 0 such that φ ( t ) K ε for all t N . Then, (23) implies that
f ( x + y ) g ( x ) h ( y ) < ε , x , y d and x y N .
Let x , y X be arbitrary elements. Since 0 < | n | < 1 for some integer n, we can choose u X such that 2 u d + N + x + y + x y . It is easy to see that:
min x u , y + u , x 2 u , y + 2 u , x y 2 u , 2 u , u d , min x 2 u , y + 2 u , x y 2 u , 2 u N .
Applying (24) and the strong triangle inequality, we obtain:
f ( x + y ) f ( x ) f ( y ) + f ( 0 ) max { g ( x u ) + h ( u ) f ( x ) , g ( u ) + h ( y + u ) f ( y ) , f ( x + y ) g ( x u ) h ( y + u ) , f ( 0 ) g ( u ) h ( u ) } < ε
for all x , y X . Since ε > 0 was arbitrary, this inequality implies that f ( x + y ) = f ( x ) + f ( y ) f ( 0 ) for all x , y X . Then, f f ( 0 ) is additive, and so f is affine. □
Remark 2.
Based on the argument presented above, it can be concluded that the assertion of Theorem 6 is still true by considering:
lim sup min { x , y } φ max { x , y } f ( x + y ) g ( x ) h ( y ) <
instead of (22).

4. Functional Inequalities

We assume that X and Y are non-Archimedean normed spaces over a non-Archimedean field F with a valuation | . | .
Lemma 1.
Let 0 < | k | < 1 for some integer k, and f : X Y with f ( 0 ) = 0 satisfy:
f ( x + y ) f ( x ) f ( y ) f ( x y ) f ( x ) + f ( y ) , x , y X .
Then, f is rational homogeneous, i.e., f ( r x ) = r f ( x ) for all x X and r Q .
Proof. 
Letting y = x in (25) and applying f ( 0 ) = 0 , we obtain f ( 2 x ) = 2 f ( x ) for all x X . Now, we will prove by induction that f ( n x ) = n f ( x ) for all x X and n N . Suppose this result holds for all 1 k < n . Then, by letting x = ( n 1 ) y in (25), we obtain:
f ( n y ) ( n 1 ) f ( y ) f ( y ) ( n 2 ) f ( y ) ( n 1 ) f ( y ) + f ( y ) = 0 , y X .
Hence, f ( n y ) = n f ( y ) for all y X . Putting y = k 2 x in (25), we obtain:
f ( x ) + f ( x ) | k | 2 f ( x ) + f ( x ) , x X .
Since 0 < | k | < 1 , we infer that f ( x ) = f ( x ) for all x X . Therefore, f ( n x ) = n f ( x ) for all x X and n Z . Let x X and r = m n be a rational number. Then, n f ( r x ) = f ( n r x ) = f ( m x ) = m f ( x ) for all x X . This implies that f ( r x ) = r f ( x ) , and the proof is completed. □
The following example shows that the assumption f ( 0 ) = 0 is necessary in Lemma 1.
Example 4.
Let Q p be equipped with the p-adic norm | . | p , and f : Q p Q p be defined by f ( x ) = x + 3 for all x Q p . Then, f satisfies (25), but f is not rational homogeneous.
Theorem 7.
Let | 2 | < 1 , ε 0 and Y be a Banach non-Archimedean space. Suppose that the function f : X Y satisfies:
f ( x + y ) f ( x ) f ( y ) f ( x y ) f ( x ) + f ( y ) + ε , x , y d
for some d > 0 . Then, there exists a unique rational homogeneous mapping A : X Y such that:
f ( x ) A ( x ) f ( 0 ) + ε , x d .
Proof. 
Letting y = x in (26), we obtain:
f ( 2 x ) 2 f ( x ) f ( 0 ) + ε , x d .
Replacing x by x 2 n + 1 in the above inequality and multiplying the resulting inequality by | 2 | n yields:
2 n f x 2 n 2 n + 1 f x 2 n + 1 | 2 | n ( f ( 0 ) + ε ) , x d , n 0 .
Then:
2 n f x 2 n 2 m f x 2 m = k = m n 1 2 k + 1 f x 2 k + 1 2 k f x 2 k max 2 k + 1 f x 2 k + 1 2 k f x 2 k : m k n 1 | 2 | m ( f ( 0 ) + ε ) ,
for all x d and n > m 0 . It follows from (28) that the sequence 2 n f x 2 n n = 1 is a Cauchy sequence for all x X . Since Y is a Banach non-Archimedean space, the sequence 2 n f x 2 n n = 1 is convergent. We define:
A : X Y , A ( x ) : = lim n 2 n f x 2 n .
Since | 2 | < 1 , we obtain A ( 0 ) = 0 . By the definition of A, (26) yields:
A ( x + y ) A ( x ) A ( y ) A ( x y ) A ( x ) + A ( y ) , x , y X .
By Lemma 1, A is rational homogeneous. Taking m = 0 and letting n in (29), we obtain (27) for all x d . The uniqueness of A easily follows from (27). □
Lemma 2.
Let 0 < | k | < 1 and f : X Y with f ( 0 ) = 0 satisfy:
f x + y k f ( x ) + f ( y ) k f ( x + y ) f ( x ) f ( y ) , x , y X .
Then, f is additive.
Proof. 
Letting y = 0 in (30), we obtain: f x k = f ( x ) k for all x X . Then, (30) yields:
f ( x + y ) f ( x ) f ( y ) | k | f ( x + y ) f ( x ) f ( y ) , x , y X .
Then, f ( x + y ) = f ( x ) + f ( y ) for all x , y X because of | k | < 1 . □
Remark 3.
Let Q 2 be equipped with the two-adic norm | . | 2 . The function f : Q 2 Q 2 given by f ( x ) = 2 (for all x Q 2 ) fulfils (30) with k = 2 , but f is not additive.
For convenience, let us define:
A d : = { ( x , y ) X × Y : x d } , B d : = { ( x , y ) X × Y : y d } , C d : = { ( x , y ) X × Y : x + y d }
for some d > 0 .
Theorem 8.
Let | k | < 1 for some non-zero integer, ε 0 and Y be a Banach non-Archimedean space. Suppose that f : X Y satisfies:
f x + y k f ( x ) + f ( y ) k f ( x + y ) f ( x ) f ( y ) + ε
for all ( x , y ) A d ( r e s p e c t i v e l y , ( x , y ) B d o r ( x , y ) C d ) , where d > 0 is a constant. Then, there exists a unique additive mapping A : X Y such that:
f ( x ) A ( x ) | k | 1 | k | max f ( 0 ) + 2 ε , f ( 0 ) + ε | k | .
Proof. 
Since the both sides of (31) are symmetric in x and y, and A d C d , it is sufficient to consider (31) for all ( x , y ) A d . Letting y = 0 in (31), we obtain:
f x k f ( x ) + f ( 0 ) k f ( 0 ) + ε , x d .
Then:
k n + 1 f x k n + 1 k n f x k n | k | n f ( 0 ) + ε , x d , n 0 ,
k n + 1 f x k n + 1 k m f x k m | k | m f ( 0 ) + ε , x d , n > m 0 .
Similar to the process of proving Theorem 7, it can be concluded that the sequence k n f x k n n = 1 is convergent and the function A : X Y can be defined as follows:
A ( x ) : = lim n k n f x k n , x X .
By the definition of A and (31), we obtain A ( 0 ) = 0 and:
A x + y k A ( x ) + A ( y ) k A ( x + y ) A ( x ) A ( y ) , x , y X .
Hence, A is additive by Lemma 2. Taking m = 0 and letting n in (33), we obtain:
f ( x ) A ( x ) f ( 0 ) +   ε , x d .
To achieve the inequality (8), suppose that x is a non-zero and arbitrary element of X. According to 0 < | k | < 1 , we can choose y X such that min { y , x + y } d . Then, (34) yields:
f x + y k A x + y k f ( 0 ) +   ε , f ( y ) k A ( y ) k f ( 0 ) +   ε | k | , f ( x + y ) A ( x + y ) f ( 0 ) +   ε .
Using the strong triangle inequalities and the additivity of A, we obtain the following results from the above inequalities:
f x + y k f ( y ) k A x k f ( 0 ) +   ε | k | , f ( x + y ) f ( y ) A ( x )   f ( 0 ) +   ε .
Then:
f ( x + y ) f ( x ) f ( y ) max { f ( x + y ) f ( y ) A ( x ) , A ( x ) f ( x ) } max { f ( 0 ) + ε , A ( x ) f ( x ) } .
According to this inequality and (31), we obtain:
f ( x ) k A ( x ) k max f x + y k f ( x ) + f ( y ) k , A ( x ) k + f ( y ) k f x + y k max f ( x + y ) f ( x ) f ( y ) +   ε , f ( 0 ) +   ε | k | max f ( 0 ) + 2 ε , A ( x ) f ( x ) +   ε , f ( 0 ) +   ε | k | .
Hence:
f ( x ) A ( x ) | k | 1 | k | max f ( 0 ) + 2 ε , f ( 0 ) +   ε | k | .
Obviously, the last inequality also holds true for x = 0 , and results in A being unique. □
Corollary 5.
Let | k | < 1 be a Banach non-Archimedean space for some non-zero integer and Y. Suppose that f : X Y with f ( 0 ) = 0 satisfies one of the following asymptotic properties:
( i ) lim x + y f x + y k f ( x ) + f ( y ) k f ( x + y ) f ( x ) f ( y ) = 0 ; ( i i ) lim x f x + y k f ( x ) + f ( y ) k f ( x + y ) f ( x ) f ( y ) = 0 ; ( i i i ) lim max { x , y } f x + y k f ( x ) + f ( y ) k f ( x + y ) f ( x ) f ( y ) = 0 .
Then, f is additive.
Proof. 
Let ε > 0 be an arbitrary real number. By the assumption ( i ) , there exists d > 0 such that (31) holds. By Theorem 8, there exists an additive mapping A ε : X Y such that f ( x ) A ε ( x ) C ε for all x X , where the constant C > 0 is independent of ε . Therefore, by the strong triangle inequality and the additivity of A ε , we obtain:
f ( x + y ) f ( x ) f ( y ) max { f ( x + y ) A ε ( x + y ) , A ε ( x ) f ( x ) , f ( y ) A ε ( y ) } C ε , x , y X .
Since ε > 0 was arbitrary, we infer that f is additive. For the other cases, the proof will be the same. □

5. Conclusions

In this work, we studied the Pexider–Cauchy functional equation. We established a new strategy to study the Hyers–Ulam stability and hyperstability of the Pexider–Cauchy functional equation in the non-Archimedean normed spaces with restricted domains. We also showed that, subject to certain restrictions, every function which fulfils certain inequalities can be approximated by an additive function in non-Archimedean normed spaces. As a consequence, we used the results to investigate the asymptotic stability behaviour of the Cauchy and Pexider–Cauchy functional equations.

Author Contributions

Conceptualization, H.G., M.B.M., A.N. and J.-H.B.; methodology, H.G., M.B.M., A.N. and J.-H.B.; software, H.G., M.B.M., A.N. and J.-H.B.; validation, H.G., M.B.M., A.N. and J.-H.B.; formal analysis, H.G., M.B.M., A.N. and J.-H.B.; investigation, H.G., M.B.M., A.N. and J.-H.B.; resources, H.G., M.B.M., A.N. and J.-H.B.; data curation, H.G., M.B.M., A.N. and J.-H.B.; writing—original draft preparation, H.G., M.B.M., A.N. and J.-H.B.; writing—review and editing, H.G., M.B.M., A.N. and J.-H.B.; visualization, H.G., M.B.M., A.N. and J.-H.B.; supervision, H.G., M.B.M., A.N. and J.-H.B.; project administration, H.G., M.B.M., A.N. and J.-H.B.; funding acquisition, H.G., M.B.M., A.N. and J.-H.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Ulam, S.M. Problems in Modern Mathematics, Science Editions; John Wiley Sons, Inc.: New York, NY, USA, 1964. [Google Scholar]
  2. Hyers, D.H. On the stability of the linear functional equation. Proc. Natl. Acad. Sci. USA 1941, 27, 222–224. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  3. Aoki, T. On the stability of the linear transformation in Banach spaces. J. Math. Soc. Jpn. 1950, 2, 64–66. [Google Scholar] [CrossRef]
  4. Rassias, T.M. On the stability of the linear mapping in Banach spaces. Proc. Am. Math. Soc. 1978, 72, 297–300. [Google Scholar] [CrossRef]
  5. Găvruta, P. A generalization of the Hyers–Ulam–Rassias stability of approximately additive mappings. J. Math. Anal. Appl. 1994, 184, 431–436. [Google Scholar] [CrossRef] [Green Version]
  6. Czerwik, S. Functional Equations and Inequalities in Several Variables; World Scientific Publishing Co., Inc.: River Edge, NJ, USA, 2002. [Google Scholar]
  7. Forti, G.-L. Hyers–Ulam Stability of Functional Equations in Several Variables. Aequationes Math. 1995, 50, 143–190. [Google Scholar] [CrossRef]
  8. Hyers, D.H.; Isac, G.; Rassias, T. Stability of Functional Equations in Several Variables; Birkhäuser: Boston, MA, USA, 1998. [Google Scholar]
  9. Jung, S.-M. Hyers–Ulam–Rassias Stability of Functional Equations in Nonlinear Analysis; Springer: New York, NY, USA; Dordrecht, The Netherlands; Heidelberg, Germany; London, UK, 2011. [Google Scholar]
  10. Kannappan, P. Functional Equations and Inequalities with Applications; Springer: Berlin/Heidelberg, Germany, 2009. [Google Scholar]
  11. Khrennikov, A. Non-Archimedean Analysis, Quantum Paradoxes, Dynamical Systems and Biological Models; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1997. [Google Scholar]
  12. Arriola, L.M.; Beyer, W.A. Stability of the Cauchy functional equation over p-adic fields. Real Anal. Exch. 2006, 31, 125–132. [Google Scholar] [CrossRef]
  13. Mirmostafaee, A.K. Approximately additive mappings in non-Archimedean normed spaces. Bull. Korean Math. Soc. 2009, 46, 387–400. [Google Scholar] [CrossRef] [Green Version]
  14. Moslehian, M.S.; Rassias, T.M. Stability of functional equations in non-Archimedean spaces. Appl. Anal. Discrete Math. 2007, 1, 325–334. [Google Scholar] [CrossRef] [Green Version]
  15. Khosravi, B.; Moghimi, M.B.; Najati, A. Asymptotic aspect of Drygas, quadratic and Jensen functional equations in metric abelian groups. Acta Math. Hungar. 2018, 155, 248–265. [Google Scholar] [CrossRef]
  16. Najati, A.; Kang, J.I.; Cho, Y.J. Local stability of the Pexiderized Cauchy and Jensen’s equations in fuzzy spaces. J. Inequal. Appl. 2011, 2011, 78. [Google Scholar] [CrossRef] [Green Version]
  17. Najati, A.; Rassias, T.M. Stability of the Pexiderized Cauchy and Jensen’s equations on restricted domains. Commun. Math. Anal. 2010, 8, 125–135. [Google Scholar]
  18. Najati, A.; Cho, Y.J. Generalized Hyers–Ulam stability of the Pexiderized Cauchy functional equation in non-Archimedean spaces. Fixed Point Theory Appl. 2011, 2011, 309026. [Google Scholar] [CrossRef] [Green Version]
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Gharib, H.; Moghimi, M.B.; Najati, A.; Bae, J.-H. Asymptotic Stability of the Pexider–Cauchy Functional Equation in Non-Archimedean Spaces. Mathematics 2021, 9, 2197. https://doi.org/10.3390/math9182197

AMA Style

Gharib H, Moghimi MB, Najati A, Bae J-H. Asymptotic Stability of the Pexider–Cauchy Functional Equation in Non-Archimedean Spaces. Mathematics. 2021; 9(18):2197. https://doi.org/10.3390/math9182197

Chicago/Turabian Style

Gharib, Hamid, Mohammad B. Moghimi, Abbas Najati, and Jae-Hyeong Bae. 2021. "Asymptotic Stability of the Pexider–Cauchy Functional Equation in Non-Archimedean Spaces" Mathematics 9, no. 18: 2197. https://doi.org/10.3390/math9182197

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop