TY - JOUR AU - Gusti Ayu Putri AU - Ni Putu Hendayanti AU - Maulida Nurhidayati PY - 2017/09/27 Y2 - 2024/03/28 TI - PEMODELAN DATA DERET WAKTU DENGAN AUTOREGRESSIVE INTEGRATED MOVING AVERAGE DAN LOGISTIC SMOOTHING TRANSITION AUTOREGRESSIVE JF - Jurnal Varian JA - Varian VL - 1 IS - 1 SE - Articles DO - https://doi.org/10.30812/varian.v1i1.50 UR - https://journal.universitasbumigora.ac.id/index.php/Varian/article/view/50 AB - Time series analysis is a statistical analysis that can be applied on data related to time. Modeling of time series data is widely associated with the process of forecastinga certain characteristics in the coming period. Most inflation data modeling is done using a linear time series models such as Autoregressive Integrated Moving Average (ARIMA). In fact only the ARIMA model can be applied to models of linear time series data. Models of ARIMA hasn't been able to give good results when the data being analyzed is a nonlinear time series data. The inflation data, data that has a tendency to form patterns of nonlinear data so the application of nonlinear time series models can be done on the inflation data. Logistic model Smooting Threshold Autoregressive (LSTAR) is a time series model can be applied to data that follow nonlinearmodel. LSTAR then developed on data-financial and economic data such as inflation. If the inflation data are modelled with expected LSTAR approach can get a better result because already done smoothing in it. This research aims to know the best model that can be used to perform data modeling inflation. The results showed that the results of the comparison of the MSE and the RMSE for the model of ARIMA and LSTAR. Based on these results it is known that the model MSE has a value and LSTAR RMSE smaller compared to ARIMA. So the model more appropriate LSTAR is used to model the data of inflation ER -