A simple functional operator (Q1304158)

The authors consider a linear operator \(F\) defined for an arbitrary mapping \(f:E^{n-1}\to E_1\), where \(E\), \(E_1\) are complex vector spaces, by the formula \[ \begin{multlined} (Ff)(x_1, \dots, x_n):=(-1)^{n-1} f(x_1, \dots,x_{n-1})-f(x_2, \dots, x_n)+ \\ \sum^{n- 1}_{s=1}(-1)^{s+1} f(x_1,\dots, x_s+x_{s+1}, \dots, x_n).\end{multlined} \] The operator \(F\) has the property \(F\circ F=0\). The following functional equation \[ (Ff)\;(x_1, \dots, x_{n+1})= 0\tag{1} \] is solved in the space of analytic functions \(f:E^n\to E_1\). The general solution of the equation (1) is given by the formula \[ f(x_1, \dots, x_n)=(Fg)(x_1, \dots,x_n)+ L(x_1, \dots, x_n),\tag{2} \] where \(g:E^{n-1}\to E_1\) is an analytic function and \(L:E^n \to E_1\) is a linear operator. Let us note that for \(n=1\) the equation (1) becomes the ordinary Cauchy functional equation for additive mappings: \[ f(x_1+x_2) -f(x_1) -f(x_2)=0. \] If \(n=2\), then (1) can be written in the form \[ f(x_1,x_2) -f(x_2,x_3) +f(x_1+ x_2,x_3)- f(x_1,x_2+ x_3)=0. \] The method of solving the equation (1) is different from the methods presented by \textit{M. Kuczma} in his book [Functional equations in a single variable (1968; Zbl 0196.16403)].