DDNS Discrete Dynamics in Nature and Society 1607-887X 1026-0226 Hindawi 10.1155/2018/6935069 6935069 Research Article Boundedness and Asymptotic Stability for the Solution of Homogeneous Volterra Discrete Equations http://orcid.org/0000-0003-4545-6266 Messina E. 1 Vecchio A. 2 Kulenovic Mustafa R. S. 1 Dipartimento di Matematica e Applicazioni Università degli Studi di Napoli “Federico II” Via Cintia 80126 Napoli Italy unina.it 2 C.N.R. National Research Council of Italy Institute for Computational Application “Mauro Picone” Via P. Castellino 111 80131 Napoli Italy cnr.it 2018 1522018 2018 17 11 2017 09 01 2018 15 01 2018 1522018 2018 Copyright © 2018 E. Messina and A. Vecchio. 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.

We consider homogeneous linear Volterra Discrete Equations and we study the asymptotic behaviour of their solutions under hypothesis on the sign of the coefficients and of the first- and second-order differences. The results are then used to analyse the numerical stability of some classes of Volterra integrodifferential equations.

GNCS-INdAM
1. Introduction

Linear Volterra Discrete Equations (VDEs) are usually represented according to two types of formulae (see, e.g.,  and references therein, [2, XIII-10], [3, Chap. 7]):(1)xn+1-xn=gn+1+an+1xn+1+j=0n+1bn+1,jxj,n0,x0  given,(2)xn=fn+j=0ncn,jxj,n1,x0  given.Even if each of the equations above can be easily transformed into the other, we read in the literature (see, e.g., ) that (1) is the discrete analogue of a Volterra Integrodifferential Equation (VIDE), whereas (2) is seen as the discrete version of a second kind Volterra Integral Equation (VIE). This is due to the fact that the simple position(3)cn+1,n+1=an+1,cn+1,n=1+bn+1,n,cn+1,j=bn+1,j,j=0,,n-2,which transforms (2) into (1) is not meaningful when we are dealing with numerical analysis of Volterra equations.

To be more specific, a simple numerical method for the VIDE, y(t)=g(t)+A(t)y(t)+0tK(t,s)y(s)ds, has the form(4)yn+1-yn=hgtn+1+hAtn+1yn+1+h2j=0n+1wn+1,jKtn+1,tjyj,where h is the stepsize, tn=nh,wn,j are given weights, and yny(tn).

Using (3), namely, cn+1,n+1=hA(tn+1), cn+1,n=h2wn+1,nK(tn+1,tn)-1, and cn+1,j=h2wn+1,jK(tn+1,tj), (4) turns into the form of (2), the analysis of which would be complicated by the fact that the coefficients do not have the same dependence on h. For such a reason, in this paper we focus on the following homogeneous VDE:(5)xn+1-xn=an+1xn+1+j=0n+1bn+1,jxj,n0,where x0 is given and bn,j=0 for j>n, and we study its asymptotic properties exploiting its particular form.

For the sake of completeness, there is also another type of VDE widely used in literature (see, e.g., [5, 6])(6)xn+1=Anxn+j=0nBn,jxj,n0,x0  given.This is an explicit equation which can be recasted in the form (5) with bn+1,n+1=an+1=0, by imposing An=1.

Asymptotic analysis of difference equations of the form (5) or its explicit version often appeared in literature in the last decades. Some of them deal with the convolution case (bn,j=bn-j); see, for instance,  and the references therein and . Most of the known results for the nonconvolution case are based on the hypothesis of double summability of the coefficients (n=0+j=0nbn,j<+); see [1, 4, 1319]. Another interesting approach, resembling the study of continuous VIDE (see, e.g., [20, 21]), basically requires that the coefficient an+1 of (5), assumed to be negative, in some sense “prevails” on the summation of the remaining coefficients bnj. Here we would like to add another piece to the framework regarding the analysis of VDE behaviour, by considering hypotheses based on the sign of the coefficients and of their first and second differences.

Since (5) is homogeneous, it has always the trivial solution. Therefore, all the results that follow are valid automatically and no assumptions are necessary when x0=0. From now on we assume that the given datum x0 is different from zero and we want to analyse the behaviour of the corresponding solution with respect to the trivial one. In Section 2 we report our main results on the asymptotic behaviour of the nontrivial solution to (5) which are then used, in Section 3, to prove the boundedness of the solution and the convergence to zero in some cases of interest.

In the whole paper it is assumed the empty sum convention j=mMvj=0, if M<m.

2. Main Results

Let vn,j be a double-indexed sequence and define Δ1vn,j=vn+1,j-vn,j, Δ2vn,j=vn,j+1-vn,j, and Δ12vn,j=Δ1(Δ2vn,j). Our main result gives sufficient conditions for (5) to have a solution xn vanishing at infinity.

Theorem 1.

Consider (5) and assume that

n¯>0 such that 2an+bn,n0,nn¯,

bn,00,n2,Δ1bn,00,nn¯,

Δ2bn,j0,n2,jn-2,

Δ12bn,j0,nn¯1,jn-1.

Then, for any x0R, there exists x such that |xn|x,n>0. If, in addition,

2an+bn,n-a<0,n>n¯1,

or

Δ12bn,jb>0,nn¯1,jn-1,

then, for any x0R, limn+xn=0.

Proof.

Set αn=an+1xn+1+j=0n+1bn+1,jxj, then(7)xn+12-xn2=2xn+1αn-αn2,and hence(8)2xn+1αn=2xn+12an+1+bn+1,n+1+2xn+1j=0nbn+1,jxj.The second addendum in the right-hand side of (8) can be written as(9)2xn+1j=0n+1bn+1,jxj=2xn+1j=0nbn+1,jΔXj,where(10)Xn+1=j=0nxj,n0,X0=0.Applying the summation by parts rule, we have(11)2xn+1j=0nbn+1,jxj=2xn+1bn+1,n+1Xn+1-j=0nXj+1Δ2bn+1,j.By adding and subtracting Xn+1j=0nΔ2bn+1,j in the right-hand side and by setting(12)Vn,j=Xn+1-Xj+1=i=j+1nxj,jn-1,Vn,j=0  for  jn,we get(13)2xn+1j=0nbn+1,jxj=2xn+1bn+1,n+1Xn+1+j=0nVn,jΔ2bn+1,j-Xn+1j=0nΔ2bn+1,j=2xn+1j=0nVn,jΔ2bn+1,j+Xn+1bn+1,0.Now, taking into account the fact that(14)2xn+1j=mnxj=j=mn+1xj2-j=mnxj2-xn+12,0mn,we have(15)2xn+1j=0nbn+1,jxj=j=0nΔ2bn+1,jΔ1Vn,j2-xn+12+bn+1,0ΔXn+12-xn+12=j=0nΔ2bn+1,jΔ1Vn,j2-xn+12bn+1,n+1+bn+1,0ΔXn+12.By (8) and (15), (7) becomes(16)xn+12-xn2=xn+122an+1+bn+1,n+1+j=0nΔ2bn+1,jΔ1Vn,j2+bn+1,0ΔXn+12-αn2.Summing up over n, for all N>n¯0, we have(17)xN+12-x02=n=0Nxn+122an+1+bn+1,n+1+n=0Nj=0nΔ2bn+1,jΔ1Vn,j2+n=0Nbn+1,0ΔXn+12-n=0Nαn2.Now, let us consider the double summation at the right-hand side. By inverting the summation order, applying the summation by part rule and recalling that Vj,j=0, it becomes(18)n=0Nj=0nΔ2bn+1,jΔ1Vn,j2=j=0NΔ2bN+2,jVN+1,j2-n=0Nj=0nΔ12bn+1,jVn+1,j2.Taking account of this and applying the summation by part rule also to the third addendum in (17), we get(19)xN+12-x02=n=0Nxn+122an+1+bn+1,n+1+j=0NΔ2bN+2,jVN+1,j2-n=0Nj=0nΔ12bn+1,jVn+1,j2+bN+2,0XN+22-b1,0X22-n=0NΔ1bn+1,0Xn+22-n=0Nαn2.In view of the first group of hypotheses (i)–(iv), this implies(20)xN+12x02+n=0n¯-2xn+122an+1+bn+1,n+1-n=0n¯-2j=0nΔ12bn+1,jVn+1,j2-b1,0x0+x12-n=0n¯-1Δ1bn+1,0Xn+22.As the whole right-hand side does not depend on N, (20) assures the boundedness of |xn| and the first part of the theorem is proved.

In order to prove the second part of our result, let us proceed by contradiction. Assume that(21)limN+n=n¯Nxn2=+.From (19) and (20) we have xN+12C+n=n¯N+1(2an+bn,n)xn2 and in view of (v) xN+12C-an=n¯N+1xn2, which leads to an absurd because of (21). So the series in (21) converges and limn+xn2=0.

Now consider hypothesis (vbis) and once again proceed by contradiction. Assume that the series n=0+xn does not converge, therefore ϵ>0 and two increasing sequences of integers {ni} and {li} with ni>li,l0>n¯ such that j=li+1ni+1xjϵ, and hence(22)Vni+1,li2ϵ2,i0.Again, using (19) and (20), we write(23)xN+12C-n=n¯Nj=0nΔ12bn+1,jVn+1,j2.Because of (iv) we can write(24)n=n¯Nj=0nΔ12bn+1,jVn+1,j2i=0mNΔ12bni+1,liVni+1,li2,with m(N) such that nm(N)<N<nm(N)+1. This together with (22) and (vbis) leads to xN+12C-ϵ2bm(N). Since m(N)+, as N increases, this is absurd. Hence, the series n=0+xn converges and the desired result follows.

It is well known that one of the most used tools in the stability analysis of VDEs is the Lyapunov approach . As already mentioned in the introduction, among the results that can be obtained by Lyapunov techniques, the most popular are based on the hypothesis that the coefficients are summable (e.g., the result in [28, Th. 2] applied to (5) requires, among other hypotheses, that i=0n-1j=n+1+bj,i/(1-aj-bj,j)<+). Our attempt to construct new functionals for the form (5) leads inevitably to this type of hypothesis. Very few are the cases where no summability requirements are made. One of these can be found in [26, Th. 2.2], where a Lyapunov functional is constructed which allows the stability analysis of an explicit equation, provided that some conditions on the sign of the coefficients and of their Δ’s are satisfied. In order to compare the technique developed in this paper with the Lyapunov one, we refer precisely to this theorem and consider (5) with bn,n=0 and an=a<0. In this situation the hypotheses of Theorem 1 proved above guarantee that the solution vanishes also in few cases not covered by Theorem 2.2 in  (just to mention one example, the coefficient of  xn,  which should be negative in , is allowed to assume whatever sign here).

Remark 2.

It is easy to see that if hypothesis (iii) in Theorem 1 holds also for j=n-1, then (vbis) assures bnn-b, for all n>n¯. Therefore, if we assume an0, nn¯, hypothesis (vbis) becomes sufficient for (v). So (vbis) does no more represent an alternative with respect to (v) and can be dropped out. In this case Theorem 1 can be stated as follows.

Corollary 3.

Consider (5) and assume that an0, for nn¯,Δ2bn,n-10, and that (ii)–(v) hold. Then, for any x0R, limn+xn=0.

Furthermore, we point out that checking assumption (vbis) of Theorem 1 may be difficult; hence the following result can be useful.

Corollary 4.

Consider (5) and assume that (i)–(iv) hold and(25)Δ12bn,n-1b,nn¯,with b>0. Then, for any x0R, limn+xn=0.

Proof.

From (iv) and the definition of Vn,j in (12),(26)n=n¯Nj=0nΔ12bn+1,jVn+1,j2n=n¯NΔ12bn+1,nVn+1,n2=n=n¯NΔ12bn+1,nxn+12.The desired result is readily obtained by using (23) and (25).

We want to underline that Theorem 1 is strongly inspired by  where the asymptotic behaviour of a nonlinear VIDE is studied and that “in some sense” our result can be viewed as its discrete analogue. This will be illustrated in the following section.

Remark 5.

Observe that, when Δ12bn,j is of convolution type, hypothesis (iv) in Theorem 1 becomes Δ2bn0, so that the advantage of using hypothesis (iv), which allowed Δ12bn,j to have a constant sign only definitely with respect to n, is completely lost. This drawback can be overcome if we know that the sign of xn is definitely constant, as it is shown in the following theorem.

Theorem 6.

Consider (5) and assume that

n¯>0 such that xn0(<0), for nn¯,

n>n¯ such that βn=2an+2bn,n-bn,n¯0, for nnn¯,

bn,00,n0,Δ1bn,00,nnn¯,

Δ2bn,j0,n1,j<n,

p0 such that Δ12bn,j0,nn¯+p,0jn¯-2,

βn-a<0,nnn¯,

or

Δ12bn,jb>0,nn¯+p,0jn¯-2.

Then, for any x0R,limn+xn=0.

Proof.

First of all observe that (a) assures 2xn+1Vn,j0,njn¯-1. From here and (13) we derive(27)2xn+1j=0nbn+1,jxjj=0n¯-2Δ2bn+1,j2xn+1Vn,j+2xn+1Xn+1bn+1,0.Now, proceeding as in the proof of Theorem 1, we arrive to(28)xN+12-x02n=0Nβn+1xn+12+j=0n¯-2Δ2bN+2,jVN+1,j2-n=0Nj=0n¯-2Δ12bn+1,jVn+1,j2+bN+2,0XN+22-b1,0X12-n=0NΔ1bn+1,0Xn+22-n=0Nαn2,which, taking into account (b), (c), and (e), assures(29)xN+12C+n=n-1Nβn+1xn+12,or(30)xN+12C-n=n¯+p-1Nj=0n¯-2Δ12bn+1,jVn+1,j2,which corresponds to (14) and (23) of Theorem 1, respectively. The desired result follows as in the proof of Theorem 1.

As a consequence of this result, the following can be easily proved.

Corollary 7.

Under assumptions (b)–(f) of Theorem 6 a sequence xn, obtained by (5) with x0R, cannot diverge and if it is convergent then its limit is zero.

Remark 8.

If, in Theorem 6, n¯=1, then hypothesis (e) can be removed, and the theorem assumes a simplified form.

3. Examples of Applications

Consider the following theoretical examples of application of Corollaries 3 and 4.

It can be easily seen that (5), with the choices for an and bn,j(31)an=0,bnj=-e-n-j-1n+12;(32)an=12nn+1+1n+12,bnj=-e-n-j-1n+12;(33)an=-1,bnj=-e-2n-jjn-12;(34)an=-1,bnj=-e-2n-jjn-12-1,satisfies the assumptions of the corollaries in the previous section.

To be more specific (31) fulfills both corollaries with n¯=0. Equation (33) satisfies only Corollary 3 with n¯=3, whereas (32) satisfies Corollary 4 but not 2.1, because bnn+an0. Equation (34) is only a slight modification of (33) and, like (33), it fulfills all the hypotheses of Corollary 3; furthermore it can be easily seen that j=0nbnj is an unbounded sequence. So (5) with coefficients as in (34) is an example of VDE with vanishing solution and nonsummable coefficients.

As a counterexample, consider (5) with coefficients given by(35)an=1+12n+11+2n-1n+12,bnj=-e-n-j-1n+12.Here condition (i) for the coefficients an and bn,n in Theorem 1 is violated and the boundedness of the solution of (5) is not guaranteed any more. In fact, this is clear in Figure 1, which shows the actual behaviour of xn.

Problem (35): unbounded xn.

Theorem 6 can be applied to the following example:(36)an=-20,x0=-10,bn,j=-12n+1e-n+1n+12,j=0,-n+1-je-n-j+1n+12,1jn.First of all we need to show that (a) holds with n¯=2. Starting from the initial condition given in (36), by simple computation, we have x1=-0.003<0,0<x2<1. Our aim is to prove that 0<xn1, for n2. Let us proceed by induction on n. Assume 0xj1,j=2,,n-1 and verify that the same is true for xn given by(37)1+20-bn,nxn=bn,0x0+bn,1x1+j=2n-2bn,jxj+1+bn,n-1xn-1.From the definition of bn,j in (36), it easily follows that -4<1+j=0n-1bn,j<0. Then taking into account the induction hypothesis and that bn,n-1-1, we obtain(38)-4<j=0n-2bn,jxj+1+bn,n-1xn-1<0.As x0=-10 and x1<0, it turns out that the right-hand side of (37) is positive, then xn>0. On the other hand (38) implies xn<(bn,0x0+bn,1x1)/21,n2, with bn,0-1/2,bn,1-1, which assures 0<xn<1. We conclude that hypothesis (a) of Theorem 6 is satisfied. Since (d) is true and (c), (f) are obvious with n=0, it remains to prove (e). In our case (e) corresponds to p0 such that Δ12bn,j0, for n2+p and j=0. Observe that Δ12bn,0 can be written as Δ12bn,0=bn+1,1-bn,1-2(bn+1,0-bn,0)+Δ1bn,0. So Δ12bn,00,n2, since (c) holds and bn+1,1-bn,1-2(bn+1,0-bn,0)=(e-1)e-(n+1)(n(e-1)-2). In conclusion, all the hypotheses of Theorem 6 are fulfilled and xn0, as can be seen in Figure 2.

Problem (36): vanishing xn.

Remark 9.

We want explicitly to mention that Theorem 1 cannot be applied to (36) because Δ12bn,n-1=e2-4e+3/e2<0,n2, and hypothesis (iv) is not satisfied.

Finally, we observe that, if in (36) we choose bn,0 according to the remaining coefficients for j0, that is, bn,0=-((n+1)e-n+1/(n+1)2), then Theorem 6 is still valid with n¯=1. So we are in the case of Remark 8 and the assumption (e) of Theorem 6 can be ignored.

A more practical application of our results is the study of the longtime behaviour of the numerical solution to VIDEs. Let us consider the homogeneous problem(39)yt=Atyt+0tKt,sysds,t0,y0=y0,and a simple method of family (4), the Backward Euler method (see , [12, (3.8)])(40)yn+1-yn=hAtn+1yn+1+h2j=1n+1Ktn+1,tjyj,n0,where h>0 is the stepsize. With the help of the results of the previous section we can prove the following.

Theorem 10.

Consider (40) and assume that

t¯ such that A(t)0,t>t¯;

K(t,0)0,t0,K(t,0)/t0,t>t¯;

K(t,s)/s0,t>0;

2K(t,s)/ts0,t>t¯.

Then the solution yn of (40) is bounded y0R. If in addition

A(t)-A<0,t>t¯,

or

2K(t,s)/tsK>0,t>t¯,s<t¯,

then limn+yn=0.

Proof.

Note that (5) coincides with (40) whenever an=hA(tn),n1, and bnj=h2K(tn,tj),n1,jn. Now, assumptions (i)–(iv) immediately assure (i)–(iv) of Theorem 1 for any fixed h and n¯(h) such that n¯(h)h=t¯. Moreover, (v) implies (v) of Theorem 1 with a=hA, so that Corollary 3 holds. In order to exploit (vbis) note that it is equivalent to 2K(t,s)-Kts/ts>0,t>t¯,s<t¯, which in turn implies that, for any fixed h, the function Γ(t,s,h)=K(t+h,s)-K(t,s)-Kts is increasing with respect to s, so that (vbis) of Theorem 1 is fulfilled with b=h4K and n¯ given above. Once again all the hypotheses of Theorem 1 are satisfied and the desired result follows.

Remark 11.

Theorem 10 would be particularly interesting whenever it is known that, under the same hypotheses, also the analytical solution y(t) of (39) goes to zero as t tends to infinity.

As we mentioned in the previous section, the analogue of Theorem 1 in the continuous case can be obtained following the line of the proof of Theorem 1 in . The following is a reformulation of such a theorem suited to our case.

Theorem 12.

Assume that hypotheses (i)–(iv) of Theorem 10 are valid, then the solution y(t) of (39) is bounded, for any choice of y0R. Furthermore, if hypothesis (v) or (vbis) of Theorem 10 holds and(41)y  isuniformlycontinuous,then limt+y(t)=0.

Remark 13.

From Theorems 1 and 10 it is clear that the asymptotic behaviour of the solution to (39) is preserved both in a generic discretization of the kind (5), where bnj represents the samples K(tn,tj), and in the numerical solution obtained by the Backward Euler method (40).

Moreover, it is worth noting that if in (39)(42)0tKt,sds<+,then assumption (41) is automatically verified; nevertheless in the discrete case, the summability of the coefficients, which is the analogous of (42), is not required as showed in example (34).

In the literature we sometimes encounter VIDEs with the following structure (see, e.g., ):(43)yt=Atyt+0tft-sstyτdτds,t0,y0=y0.This kind of equation can be easily recast in the form (39), with K(t,s)=t-stf(τ)dτ, and the method (40) for it will read(44)yn+1-yn=hAtn+1yn+1+h2i=1n+1j=n+1-in+1ftjyj,n0.Theorem 10 immediately becomes as follows.

Corollary 14.

Consider (44) and assume that

t¯>0:A(t)0,t>t¯,

f(t)0,t0,

f(t)0,t0,

A(t)A<0,tt¯,

or

f(t)K>0,t>0.

Then, for any y0R,limn+yn=0.

A comparison to Theorem 12 reveals that, contrarily to (iv), hypothesis (a3) is required to hold in the whole integration range. This is due to the fact that (see Remark 5 in case of (43)) 2K(t-s)/ts is of convolution type. The application of Theorem 6 leads to the following result.

Theorem 15.

Consider (44) and assume that

n¯0:yn0(<0),n>n¯;

t~>0:A(t)0,t>t~;

f(t)0,t>0;

f(t)0,t>t~.

Then, for any y0R, limn+yn=0.

Proof.

We want to prove that all the hypotheses of Theorem 6 are fulfilled with an=hA(tn) and bnj=h2i=n-jnfti. Let h>0 be fixed and let n~:=n~(h) be such that n~(h)h=t~. Hypothesis (a) of Theorem 6 is obviously true because of (HA). In order to prove (b), observe that (HB) together with (HC) assure that, for nn=max{n~,n¯},(45)βn=2hAtn+2h2j=0nftj-h2j=n-n¯nftj<h2ft~0.As bn,0=h2f(tn), (HC) implies the first condition in (c), and since Δ1bn,0=h2(f(tn+1)-f(tn)), (HD) immediately assures the second condition of (c). Furthermore, since Δ2bn,j=h2f(tn-j-1), (HC) also implies (d) and, taking into account the fact that Δ1,2bn+1,j=h2(f(tn+1-j)-f(tn-j)), we have that (HD) implies (e) with p=max{0,n~-2}. Finally, whenever f is not identically zero, we can always assume, with no loss of generality, that t~ is such that f(t~)<0, so that (45) assures (f) with A=-h2f(t~), and the proof is complete.

As already mentioned above, in  equation (3.32) represents the velocity of the centre of mass of a system of N particles in collective motion under alignment and chemotaxis effect. Here A(t)=0 and(46)ft=-c1e-c2te-c3/tt2,with c1,c2, and c3 depending on the number of particles, their dimension, and the dynamic of the motion. For significative values of c1,c2, and c3 hypotheses (a1), (a2), (a3), and (a4bis) of Theorem 15 are satisfied. Hence, we expect that in the numerical simulation of (43) obtained by using the Backward Euler method (44), any convergent numerical solution vanishes at infinity. Figure 3 shows exactly this behaviour when integrating (43)–(46) with c1=105,c2=10-2, and c3=10-4.

Problem (43)–(46): vanishing yn.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

Acknowledgments

The research was supported by GNCS-INdAM.

Gyori I. Reynolds D. W. Sharp conditions for boundedness in linear discrete Volterra equations Journal of Difference Equations and Applications 2009 15 11-12 1151 1164 10.1080/10236190902932726 MR2569138 Feller W. An Introduction to Probability Theory and ITS Applications 2008 New York, NY, USA John Wiley & Sons MR0038583 Linz P. Analytical and Numerical Methods for Volterra Equations 1985 Philadelphia, Pa, USA SIAM Studies in Applied Mathematics 10.1137/1.9781611970852 MR796318 Berezansky L. Migda M. Schmeidel E. Some stability conditions for scalar Volterra difference equations Opuscula Mathematica 2016 36 4 459 470 10.7494/OpMath.2016.36.4.459 MR3488501 Elaydi S. Murakami S. Uniform asymptotic stability in linear Volterra difference equations Journal of Difference Equations and Applications 1998 3 3-4 203 218 MR1616085 10.1080/10236199808808097 Elaydi S. An Introduction to Difference Equations 2005 New York, NY, USA Springer MR2128146 Bohner M. Sultana N. Subexponential solutions of linear Volterra difference equations Nonautonomous Dynamical Systems 2015 2 63 76 MR3412180 Cuevas C. Choquehuanca M. Soto H. Asymptotic analysis for Volterra difference equations Asymptotic Analysis 2014 88 3 125 164 MR3225656 2-s2.0-84905052908 10.3233/ASY-131213 Elaydi S. Messina E. Vecchio A. On the asymptotic stability of linear Volterra difference equations of convolution type Journal of Difference Equations and Applications 2007 13 12 1079 1084 MR2364952 10.1080/10236190701264529 2-s2.0-36248934697 Elaydi S. Murakami S. Asymptotic stability versus exponential stability in linear Volterra difference equations of convolution type Journal of Difference Equations and Applications 1996 2 4 401 410 MR1444269 10.1080/10236199608808074 Medina R. Asymptotic behavior of Volterra difference equations Computers & Mathematics with Applications 2001 41 5-6 679 687 10.1016/S0898-1221(00)00312-6 MR1822595 Vecchio A. Stability results on some direct quadrature methods for Volterra integro-differential equations Dynamic Systems and Applications 1998 7 4 501 518 MR1664586 Choi S. K. Koo N. J. Asymptotic property of linear Volterra difference systems Journal of Mathematical Analysis and Applications 2006 321 1 260 272 MR2236556 10.1016/j.jmaa.2005.08.028 2-s2.0-33646351525 Diblik J. Ruzickova M. Schmeidel E. Existence of asymptotically periodic solutions of scalar Volterra difference equations Tatra Mountains Mathematical Publications 2009 43 51 61 10.2478/v10127-009-0024-7 MR2588875 Diblik J. Schmeidel E. On the existence of solutions of linear Volterra difference equations asymptotically equivalent to a given sequence Applied Mathematics and Computation 2012 218 18 9310 9320 10.1016/j.amc.2012.03.010 MR2923029 Elaydi S. Stability and asymptoticity of Volterra difference equations: a progress report Journal of Computational and Applied Mathematics 2009 228 2 504 513 MR2523167 10.1016/j.cam.2008.03.023 2-s2.0-64549151631 Kolmanovskii V. Shaikhet L. Some conditions for boundedness of solutions of difference volterra equations Applied Mathematics Letters 2003 16 6 857 862 2-s2.0-0141684888 10.1016/S0893-9659(03)90008-5 Migda M. Růžičková M. Schmeidel E. Boundedness and stability of discrete Volterra equations Advances in Difference Equations 2015 2015, article 47 1 11 10.1186/s13662-015-0361-6 MR3314614 Song Y. Baker C. T. Perturbations of Volterra difference equations Journal of Difference Equations and Applications 2004 10 4 379 397 MR2047218 10.1080/10236190310001625253 2-s2.0-1842511628 Burton T. A. Volterra Integral and Differential Equations 2005 202 Amsterdam, The Netherlands Elsevier Mathematics in Science and Engineering MR2155102 Zbl1075.45001 Crisci M. R. Kolmanovskii V. B. Russo E. Vecchio A. Boundedness of discrete Volterra equations Journal of Integral Equations and Applications 1997 211 1 106 130 10.1006/jmaa.1997.5430 Adivar M. Raffoul Y. N. Necessary and sufficient conditions for uniform stability of Volterra integro-dynamic equations using new resolvent equation Analele Ştiinţifice ale Universităţii Ovidius Constanţa. Seria Matematică 2013 21 3 17 32 2-s2.0-84888621088 10.2478/auom-2013-0039 Adivar M. Raffoul Y. N. Qualitative analysis of nonlinear Volterra integral equations on time scales using resolvent and Lyapunov functionals Applied Mathematics and Computation 2016 273 258 266 10.1016/j.amc.2015.09.087 MR3427751 Crisci M. R. Kolmanovskii V. B. Russo E. Vecchio A. Stability of continuous and discrete Volterra integro-differential equations by Liapunov approach Journal of Integral Equations and Applications 1995 7 4 393 411 MR1382061 10.1216/jiea/1181075895 2-s2.0-0001074423 Crisci M. R. Kolmanovskii V. B. Russo E. Vecchio A. Stability of difference Volterra equations: direct Liapunov method and numerical procedure Computers & Mathematics with Applications 1998 36 10-12 77 97 MR1666128 Crisci M. R. Kolmanovskii V. B. Russo E. Vecchio A. Stability of discrete Volterra equations of Hammerstein type Journal of Difference Equations and Applications 2000 6 2 127 145 MR1760154 10.1080/10236190008808218 2-s2.0-0013106905 Kublik C. Raffoul Y. Lyapunov functionals that lead to exponential stability and instability in finite delay Volterra difference equations Acta Mathematica Vietnamica 2016 41 1 77 89 MR3452956 10.1007/s40306-014-0098-4 2-s2.0-84955489871 Raffoul Y. N. Boundedness and periodicity of Volterra systems of difference equations Journal of Difference Equations and Applications 1998 4 4 381 393 MR1657161 10.1080/10236199808808150 Kiffe T. R. On nonlinear Volterra equations of nonconvolution type Journal of Differential Equations 1976 22 2 349 367 MR0417721 10.1016/0022-0396(76)90033-4 2-s2.0-49549136447 Brunner H. van der Houwen P. J. The Numerical Solution of Volterra Equations 1986 Amsterdam, The Netherlands North-Holland Publishing Co. Di Costanzo E. Natalini R. A hybrid mathematical model of collective motion under alignment and chemotaxis https://arxiv.org/abs/1507.02980