ABSTRACT

The main objective of the study is to construct a mathematical model which could be used to produce dependable estimates of future values of production. The data for this study is a secondary data on production collected from Nigerian Bottling Company PLC Aba plant. The data is presented on monthly basis in millions and covers the period of January 1993 to December 2001. The chi-square goodness — of— fit test was used to test the hypothesis that the data were from a normal population. The data were evaluated for constant variance using the Bartlett's test. Runs up and down was used to test for randomness. From the plot of the original data, it showed neither an upward nor downward movement which implies a constant trend, then the grand mean was used as an estimate of the trend. The series was detrended, deseasonalised,autocorrelations and partial; autocorrelations of the residuals(et) were computed for the additive and multiplicative models .also, the assumption of randomness was justified by subjecting the residual(et) to serial correlation test using the auto correlations at iags(k=1 and 12) At the end, the production values for the months of the year 2002 were predicted using both models.