On a result of Gross and Yang (Q750667): Difference between revisions

Let \(S_ 1=\{a_ 1,a_ 2\}\) and \(S_ 2=\{b_ 1,b_ 2\}\) be two pair of distinct elements with \(a_ 1+a_ 2=b_ 1+b_ 2\) but \(a_ 1a_ 2\neq b_ 1b_ 2\). Suppose that there are two nonconstant entire functions f and g of finite order such that \(E_ f(s_ j)=E_ g(s_ j)\) for \(j=1,2.\) Then either \(f\equiv g\), \(f+g=a_ 1+a_ 2\), \((f-c/2)(g-c/2)\equiv \pm ((a_ 1-a_ 2)/2)^ 2\), where \(c=a_ 1+a_ 2\). This occurs only for \[ (a_ 1-a_ 2)^ 2+(b_ 1-b_ 2)^ 2=0),\quad (f-a_ j)(g-a_ k)=(-1)^{j+k}(a_ 1-a_ 2)^ 2\text{ for } j,k=1,2. \] This occurs only for \[ 3(a_ 1-a_ 2)^ 2+(b_ 1-b_ 2)^ 2=0,\quad (f-b_ j)(g-b_ k)\equiv (-1)^{j+k}(b_ 1-b_ 2)^ 2\text{ for } j,k=1,2. \] This occurs only for \((a_ 1-a_ 2)^ 2+3(b_ 1-b_ 2)^ 2=0\). This is a typical result of the author. This improves upon earlier results of the reviewer and Yang.