Abstract
In this paper one of the Box- Jenkins time series models has been applied to forecast the average of the monthly sales in the period 2003 - 2009 in Kirkuk Industry Cement.
From the estimated autocorrelation and partial autocorrelation coefficients it found that the time series of sales are nearly stationary and contains the effect of seasonality.
To select the best model using Minimum value for Mean Squared Error Criterion (Mse) & Minimum value for Akaikes Information Criterion (AIC). It appear that the Model ARIMA (2،1،2), is applicable for predict the monthly sales for the period until 2013.
Keywords
)