A Weak Law of Large Numbers Under Weak Mixing

This paper presents a new weak law of large numbers (WLLN) for heterogenous dependent processes and arrays. The dependence requirements are notably weaker than the best available current results (due to Andrews (1988)). Specifically, we show that the WLLN holds when the process is weak mixing, only requiring that the mixing coeffi cients Cesàro sum to zero. This is weaker than the conventional assumption of strong mixing. ∗Research supported by the National Science Foundation and the Phipps Chair. †Department of Economics, 1180 Observatory Drive, University of Wisconsin, Madison, WI 53706.