Weakly pinned random walk on the wall: pathwise descriptions of the phase transition. (Q1766019)

The authors consider a one-dimensional random walk which is conditioned to stay non-negative and is ``weakly pinned'' to zero. That means there is a symmetric, nearest neighbor random walk \(\{S_n\}_{n\geq 0}\) on \(Z\). For \(n\geq 0\) and \(x\in Z\), the number of \(0\leq j\leq n-1\) such that \(S_j=x\) is denoted by \(L_n^x\). Then the behaviour of the random walk under the probability measure \(\mu _n^x\) is studied where \[ \mu _n^x(d\omega ) = {1\over Z_n^x }P^x[\exp (\beta L_n^0): d\omega \cap \{S_1\geq 0,\dots ,S_1\geq 0\}] \] where \(\beta \in R\) and \(Z_n^x\) is the normalizing constant. In the case \(\beta > 0\), which we are mainly interested in, the basic feature of the measure \(\mu _n^x\) can be explained as follows. The exponential factor in the formula is referred as the weak pinning which tries to localize the path of the random walk to the origin. The effect of the weak pinning, however, has to compete with the entropic fluctuation of the random walk, which together with the restriction to the non-negative values, has a tendency to push the path away from the origin. The behaviour then substantially depends on the pinning strength expressed by the parameter \(\beta \). If the pinning is less than (resp.\ equal to) the critical strength, then the limit process is the Brownian meander (resp.\ reflecting Brownian motion). If the pinning strength is supercritical, then the limiting process is positively recurrent Markov chain with a strong mixing property.