PEMODELAN TIME SERIES DAN PERAMALAN
MENGGUNAKAN METODE AUTOREGRESSIVE
MOVING AVERAGE (ARIMA) DAN RANDOM WALK
Total View This Week0
Total View This Week0
Institusion
Universitas Lampung
Author
Faiga Kharimah, 1117031022
Subject
QA Mathematics
Datestamp
2015-06-24 07:33:54
Abstract :
ABSTRACT Autoregressive integrated moving average (ARIMA) is a combination of Autoregressive (AR) model and Moving Average (MA) model. The ARIMA model has orde (0,1,0) is called RandomWalk model. The ARIMA model using past and present value to produce short-term forecasting. The purpose of this research is to determine the best ARIMA model for forecasting the Consumer Price Index (CPI) and health comodities price index Bandar Lampung city in the period January to June 2014. The ARIMA model has assumption that the series data are stationary. The CPI and health comodities price index of Bandar Lampung is not stationary, then we apllied differencing to make the data stationary. To find the best model ARIMA, first we check the stationary data by using time series plot, Autocorrelation Function (ACF), and unitroot test. Then the time series model was found by using ACF and Partial Autocorrelations Function (PACF). The best model was found by using criteria Mean Square Error (MSE), Akaike?s Information Criterion (AIC) and Bayesian Information Criterion (BIC). The best model is ARIMA (1,1,0) for CPI and ARIMA (0,1,0) for health comodities price index. Key Word : time series, forecasting, CPI, ARIMA
ABSTRACT Autoregressive integrated moving average (ARIMA) is a combination of Autoregressive (AR) model and Moving Average (MA) model. The ARIMA model has orde (0,1,0) is called RandomWalk model. The ARIMA model using past and present value to produce short-term forecasting. The purpose of this research is to determine the best ARIMA model for forecasting the Consumer Price Index (CPI) and health comodities price index Bandar Lampung city in the period January to June 2014. The ARIMA model has assumption that the series data are stationary. The CPI and health comodities price index of Bandar Lampung is not stationary, then we apllied differencing to make the data stationary. To find the best model ARIMA, first we check the stationary data by using time series plot, Autocorrelation Function (ACF), and unitroot test. Then the time series model was found by using ACF and Partial Autocorrelations Function (PACF). The best model was found by using criteria Mean Square Error (MSE), Akaike?s Information Criterion (AIC) and Bayesian Information Criterion (BIC). The best model is ARIMA (1,1,0) for CPI and ARIMA (0,1,0) for health comodities price index. Key Word : time series, forecasting, CPI, ARIMA
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
