No label defined (Q3972582)

[For the entire collection see Zbl 0685.00020.] For the class of Gaussian Markov sequences \(\{\xi_ n,\;n\geq1\}\) determined by the restrictions on the sequence of variances and that of correlation coefficients necessary and sufficient conditions are found for almost sure convergence of \(\xi_ n\) to 0. The stochastic recurrence equation \[ X_{n+1}=(1-a_{n+1})X_ n+b_{n+1}\gamma_{n+1},\qquad n\geq 0,\quad X_ 0=0, \] is considered with real numbers \(a_ n\neq 0\), \(b_ n\neq 0\) and independent standard normal random variables \(\gamma_ n\). On the base of the obtained convergence criteria necessary and sufficient conditions for \(\{a_ n\}\) and \(\{b_ n\}\) are given for almost sure convergence of \(X_ n\) to 0.