Sequential Estimation of the Parameter of a Stochastic Difference Equation with Random Coefficients (Q4030693)

The statistical data \(x= (x_ k)_{k\geq 1}\) under consideration form a random sequence that satisfies a first-order stochastic difference equation \[ x_ k= \lambda_ k x_{k-1}+ \varepsilon_ k, \qquad x_ 0=0, \] where \((\lambda_ k)_{k\geq 1}\) and \((\varepsilon_ k)_{k \geq 1}\) are two independent sequences in a probability space \((\Omega,{\mathcal F},\mathbb{P})\). It is assumed that \((\lambda_ k)_{k \geq 1}\) is a sequence of independent and identically distributed random variables with \({\mathbf E} \lambda_ k=\lambda\), \({\mathbf D} \lambda_ k= \sigma^ 2\). It is also assumed that \((\varepsilon_ k)_{k \geq 1}\) is a sequence of independent identically distributed random variables with \({\mathbf E} \varepsilon_ k=0\), \({\mathbf D} \varepsilon_ k=1\). The purpose of the paper is to construct and explore ``successive'' estimates (based on the observations \(x_ 1,x_ 2,\dots\)) of the unknown parameter \(\lambda\) under the assumption that the parameter \(\sigma\geq 0\) is known.