“…The fact that stochastic harvesting translates into pop- ulation variability has a very important implication when we analyze time series data on abundance from exploited populations. Recently, Becerra-Muñuz et al (1999) found that ARIMA models, a family of linear time series models, were less capable of explaining data from exploited populations than from unexploited ones. Our results are in line with that empirical finding, indicating that models not including harvesting may be less capable of explaining the population dynamics of exploited populations.…”