Conditional moments and linear regression for stable random variables (Q1180183)

Jointly \(\alpha\)-stable random variables with index \(0<\alpha<2\) have only finite moments of order less then \(\alpha\), but their conditional moments can be higher than \(\alpha\). We provide conditions for this to happen and use the existence of conditional moments to study the regression \(E(X_ 2\mid X_ 1=x)\). We show that if \((X_ 1,X_ 2)\) is a symmetric \(\alpha\)- stable random vector, then under appropriate conditions, the regression is well-defined even when \(\alpha\leq 1\) and linear in \(x\). The results are applied to different classes of symmetric \(\alpha\)-stable processes (authors' summary).