Conditional limit results for type I polar distributions (Q626293)

Bivariate polar random vectors \((X,Y)\) of the form \(X=R\cos\Theta\), \(Y+\rho X+\sqrt{1-\rho^2}R\sin \Theta\) are considered, where the associated random radius \(R>0\) and the random angle \(\Theta\) are independent, and \(\rho\in[-1,1]\) is a pseudocorrelation. The author derives asymptotic expansions for the conditional probability \(\bar I_{\rho,u}(y)=\text{P}\{Y>y\;| \;X>u\}\) as \(u\) tends to the upper endpoint of the distribution of \(R\). The case when the distribution of \(\Theta\) depends on \(u\) is also considered. These results are applied to the analysis of the asymptotics of unconditional probabilities \(\text{P}\{Y>y,X>x\}\) and to the estimation of the probability \(1-\bar I_{\rho,u}(y)\).