Time Series Analysis Of Monthly Rainfall (Mm) In Umudike From 1999 To 2005

ABSTRACT

 The data used in this study was collected from the Agrometeorology Department of the National Root Crops Research Institute, Umudike (NRCRI). The method of analysis of the data was time series analysis, MINITAB was employed to fit model that was capable of explaining some properties of the original data. In doing this, the original data was made stationary. Stationary time series implies that the series contained no trend and seasonal variation. These two components were removed by differencing the data. Thereafter, based on the diagnostic tools the ACF, Autocorrelation function and PACF, partial Autocorrelation function graphs (correlograms), the appropriate ARIMA Model (Autoregressive integrated moving Average), was identified. The identified model was used to forecast for the year 2006. Finally, based on the result of the analysis conclusion were drawn and recommendations were made which can be of great benefits to the farmers in Umudike, Government and Health Officers.