ScholarLens Back to search
2015•Unpublished venueRequires access

TIME SERIES MODELING AND FORECASTING OF THE CONSUMER PRICE INDEX BANDAR LAMPUNG

Faiga Kharimah, Mustofa Usman, Widiarti Widiarti, Faiz Ahmed Mohamed Elfaki

Open publisher page 10 citations

Abstract

The aims of this study are to find the best Time Series model for forecasting the Consumer Price Index (CPI). To find the best model, first we evaluate the stationary of the data by using time series plot, Autocorrelation Function (ACF), and Unit root Test. Then the Time Series model was found by using ACF and Partial Autocorrelations Function (PACF). The best model was found by using the criteria: Mean Squares Error (MSE), Akaike Information Criteria (AIC) and Bayesian Information Criterion (BIC. Based on this criteria the best modelfound in this paper is ARIMA (1,1,0) compare to ARIMA (0,1,1), and ARIMA(1,1,1).

About this research paper

What this paper is about

The aims of this study are to find the best Time Series model for forecasting the Consumer Price Index (CPI). To find the best model, first we evaluate the stationary of the data by using time series plot, Autocorrelation Function (ACF), and Unit root Test. Then the Time Series model was found by using ACF and Partial Autocorrelations Function (PACF). The best model was found by using the criteria: Mean Squares Error (MSE), Akaike Information Criteria (AIC) and Bayesian Information Criterion (BIC. Based on this criteria the best modelfound in this paper is ARIMA (1,1,0) compare to ARIMA (0,1,1), and ARIMA(1,1,1).

Why it matters

OpenAlex reports 10 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

The aims of this study are to find the best Time Series model for forecasting the Consumer Price Index (CPI). To find the best model, first we evaluate the stationary of the data by using time series plot, Autocorrelation Function (ACF), and Unit root Test. Then the Time Series model was found by using ACF and Partial Autocorrelations Function (PACF). The best model was found by using the criteria: Mean Squares Error (MSE), Akaike Information Criteria (AIC) and Bayesian Information Criterion (BIC. Based on this criteria the best modelfound in this paper is ARIMA (1,1,0) compare to ARIMA (0,1,1), and ARIMA(1,1,1).

Key concepts: Autoregressive integrated moving average, Akaike information criterion, Bayesian information criterion, Partial autocorrelation function, Box–Jenkins, Mathematics, Statistics, Autocorrelation

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
TIME SERIES MODELING AND FORECASTING OF THE CONSUMER PRICE INDEX BANDAR LAMPUNG — Research Paper | ScholarLens