This study examined the Nigerian Ports Authority (NPA) revenue generated monthly series spanning January, 2007 to December, 2019 inclusive. The objective of the study is to define a good model and a perfect fit for the NPA revenue generated series by comparing the forecast of the transformed series with outlier and without outliers using the forecast evaluation criterion. Bartlett’s power transformation technique was employed. Outlier detection methods initiated were: Modified Z score test, Median Absolute Deviation (MAD), Standard deviation, Range test and Tukey methods. Outliers were treated using the mean imputation method. Box and Jenkins technique was employed for model building. The results showed that (i) normality and variance stability was achieved (ii) MAD, Range and standard deviation tests proved effective in outlier detection (iii) ARIMA (1, 1, 3) model, perfectly described the behaviour of the NPA revenue generated series for the transformed series without outliers. (iv) The forecast evaluation criterion of the transformed data without outliers has a perfect fit when matched with the transformed data with outliers hence, ARIMA (1, 1, 3) is the most suitable model