Law of large numbers for random walk with unbounded jumps and birth and death process with bounded jumps in random environment (Q1661577)

The paper consists of two parts. In the first one, a law of large numbers is proved for a (discrete-time) random walk with unbounded steps in a stationary, ergodic, uniformly elliptic and polynomially decaying random environment. The approach is that of ``the environment viewed from the particle'' introduced by \textit{S. M. Kozlov} [Russ. Math. Surv. 40, No. 2, 73--145 (1985; Zbl 0615.60063); translation from Usp. Mat. Nauk 40, No. 2, 61--120 (1985)]. The assumptions resemble those of \textit{F. Comets} and \textit{S. Popov} [Ann. Inst. Henri Poincaré, Probab. Stat. 48, No. 3, 721--744 (2012; Zbl 1247.60139)]. The speed of decay is assumed to be higher in order to utilize a large deviation result for martingales of \textit{E. Lesigne} and \textit{D. Volný} [Stochastic Processes Appl. 96, No. 1, 143--159 (2001; Zbl 1059.60033)] and instead of assuming the random walk to be uniformly transient to the right, it is required that the annealed mean of the first hitting time of the lattice of the positive half line is finite. Except for a special case, the question of whether or when the limit of the law proved is positive, is left open.\par In the second part of the paper, the law of large numbers is proved for a continuous-time version of the process with bounded jumps such that the discrete-time skeletons are random walks with unbounded jumps in a random environment satisfying a uniform ellipticity condition. Using the latter, it is shown that the tail probability of the jumps of the skeleton process, given sufficiently small time steps, decays exponentially, so that the law of large numbers is easily obtained from the result for the discrete-time process. An explicit expression for the limit is given, and the question of when it is positive and when not, is settled.