IMPLEMENTATION OF ARIMA MODELS ON DEMAND FORECASTING AT PT WORLD YAMATEX SPINNING MILLS

No Thumbnail Available
Date
2009-12-10
Journal Title
Journal ISSN
Volume Title
Publisher
Universitas Widyatama
Abstract
Company activities are in uncertain circumtances so need for tool or method to predict or forecast the future is deeply needed. This research, forecasting method, uses Autoregressive Integrated Moving Average (ARIMA). ARIMA have ability to solve the forecasting problem by applying Autocorrelation And Partial Autocorrelation Coefficients. Temporary model are (2,2,1)(1,1,1)15, (1,1,2)(1,1,1)15, and (2,1,2)(1,1,1)15. Checking for Ljung-Box Q Statistics by Chi-Square (X2), one of the result are p-value at lag 12 is 0.01, 0.039, and 0.024 for each model, because the closest value to 0.05, so the chosen model is ARIMA(1,1,2)(1,1,1)15
Description
Keywords
Autoregressive Integrated Moving Average, Autocorrelation And Partial Autocorrelation Coefficients, Model Analysis
Citation