Time Series Model and Analysis On Rainfall In Oshogbo Osun State, Nigeria

Alawaye A.I, Alao A. N

This research work  statistical analysis of rainfall  is aim  to fit an appropriate model on the time series data on monthly distribution of rainfall in Osogbo Metropolis , to describe the behavior of rainfall on a monthly basis and to determine the variations that characterized rainfall data. The data on the monthly amount of rainfall between (2004-2015), were collected from Nigeria Meteorological Agency. The data were then subjected to time series analysis. The first step was to obtain the time plot in order to describe the observations under study. The time plot reveals that the rainfall data show high level of volatility characterized  by seasonal and irregular variations. The series moves up and down.In order to remove the seasonal variation, the researchers  subjected the data to seasonal decomposition or deseasonalization. The seasonal adjusted data were then used to plot Autocorrelation function plot method (ACF) and Partial Autocorrelation function plot (PACF). This series show a more stable series and the data was then model. The logistic model shows to be better and then used to forecast the rainfal for the next 2 years.