Abstract
Most recorded data of continuous distributions are rounded to the nearest decimal place due to the precision of the recording mechanism. This rounding entails errors in estimation and measurement. In this study, we consider parameter estimation of time series models based on rounded data. The adjusted maximum likelihood estimates in [Stam, A.-Cogger, K. O.: Rounding errors in autoregressive processes, Internat. J. Forecast. 9 (1993), 487-508] are derived theoretically for the first order moving average MA(1) model. Simulations are performed to compare the efficiencies of the adjusted maximum likelihood estimators with other estimators.