Long runs under a conditional limit distribution (Q473155)

The authors provide a sharp approximation of the density of long runs of a random walk conditioned on its end value or by an average of a function of its summands as their number tends to \(\infty\). In the large deviation range of the conditioning event, it extends the Gibbs conditional principle, in the sense that it provides a description of the distribution of the random walk on long subsequences. The authors also obtain an approximation of the density of the runs in the case in which the conditioning event states that the end value of the random walk belongs to a thin or a thick set with a nonempty interior. Such approximations hold either in probability under the conditional distribution of the random walk, or in total variation norm between measures. The authors provide an application of the described approximation scheme to the evaluation of rare event probabilities through importance sampling. In the case in which the conditioning event is in the range of the central limit theorem, it provides a tool for statistical inference, in the sense that it provides an efficient way to implement the Rao-Blackwell-Kolmogorov theorem for the improvement of estimators, and it also leads to conditional inference procedures in models with nuisance parameters. The paper is concluded by an algorithm for the simulation of such long runs, together with an algorithm determining the maximal length for which the approximation is valid up to a prescribed accuracy.