The stationary non-negative Markov chains {Yn} and {Xn} specified by the relations YnÅ+1=min(Yn,ηn)/ρ (0<ρ<1) for {ηn} a sequence of independent identically distributed (i.i.d.) random variables which are independent of {Yn}, and XnÅ+1=ρXn+ξn (0<ρ<1) for {ξn} a sequence of i.i.d. random variables which are independent of {Xn}, are mutually time-reversed if and only if their common marginal distribution is exponential, relating the exponential autoregressive process of Gaver and Lewis to the exponential minification process of Tavares.
