Characterization of the geometric distribution via linear combinations of observations and of records (Q6133737)

Let \(\{X_n\}\) be a sequence of independent identically distributed random variables and let \(\{X^{(n)}\}\) be the sequence of upper records: \(X^{(1)}=X_1\), while the second record \(X^{(2)}\) is the first subsequent variable in the sequence \(\{X_n\}\) that exceeds \(X_1\). The authors prove that, if \(X^{(1)}+X^{(2)}=X_1+2X_2\), then there exists \(p\in\left]0,1\right[\) such that \(P(X=x)=p\,(1-p)^{x-1}\) for \(x=1,2,\dots\). The converse is known. The authors also discuss the following unproved conjecture: if for some positive integer \(m>2\) \(\sum_{i=1}^m X^{(i)}=\sum_{i=1}^m i\,X_i\), then there exists \(p\in\left]0,1\right[\) such that \(P(X=x)=p\,(1-p)^{x-1}\) for \(x=1,2,\dots\)