Parameter estimation is the most important part in modelling and predictingtime series. However, the existence of outliers in the datawill affect theestimation, which consequently jeopardizesthe validity of the model. Therefore,the existence of outliers in the data must be first detected before the next processcan be performed.The best outlier detection procedurecan ensure data are free of outliersand achieve the best value parameter estimation. One of the procedures is usingthe bootstrap methodto obtain the variance of the estimated magnitude of outlier effects. The variance found directly from the bootstrap method is called the 'standard' variance. However, the bootstrap method is quitecomplexin obtainingthe variancevalue. As alternatives, trimming methodsinvolving robust estimatorssuch as amedianabsolute deviation (MADn) andalternative median-based deviation called Tnin the 'robust' variance calculation are used to replace the 'standard' variance. This method involves direct calculation to obtain the value of the variance from the estimated magnitude of outlier effects. To see the effectiveness of this method, the bilinear (1,0,1,1) modeland two robust detection procedures,namely,modified one-step M-estimator (MOM)with MADn and MOM with Tnwere used. Later, these two procedures are evaluated and compared with the bootstrap method through simulation studies based on the probability of outlier detection. Through the findings obtained, in general, the standard bootstrap procedure performsbetter than the robustprocedure performance in detecting the existence of outliers in the bilinear (1,0,1,1) model.
