Your Cart

Your cart is empty

Home  >  Volume 34

Model for Forecasting Rainfall in Nigeria by Obisesan K.O (pages 167- 176)
Sale price: $5.00
Quantity:

Abstract

Forecasting rainfall has only been done with mere comparison with cases from different locations. Much has not been done in obtaining appropriate statistical model in evaluating rainfall events in Nigeria.

In this work, Seasonal Autoregressive Integrated Moving Average (SARIMA)is used by considering monthly data for six stations: Oyo, Enugu, Kwara, Kaduna, Rivers and Tarabafor the period 1994-2014. Six SARIMA models were developed for the abovenamed stations as follows:(1,0,3)x(1,0,1)12, (3,0,3)x(1,0,1)12, (1,0,1)x(1,1,1)12,(1,0,1)x(2,0,1)12, (1,0,2)x(1,0,1)12, (3,0,3)x(1,0,1)12 respectively.

Basic exploratory data analysis was carried out, the normality of the series was also tested, the time plot of the monthly rainfall in each station was checked for trend and seasonal pattern in the series, the timeplot shows no obvious trend or seasonality, the Augmented Dickey-Fullertest, Kwaiatkowski-Phillips-Schmidt-Shin test, Phillips-Perron Unit RootTest were used to establish the stationarity in the series and Hylleberg, Engle, Granger and Yoo(HEGY's Test) was used to test the seasonal unit rootin the series, the test shows that all but Kwara series were seasonally stationary. The Kwara series was seasonally differenced. The ACF and PACFplot were used to identify the possible order of the models, while the modelswith the lowest Akaike Information Criterion (AIC) were selected. Severaldiagnostic statistics and plots of the residuals were used to examine thegoodness of fit of the selected model to the data.

These models were used to forecast the Monthly rainfall series for the fouryears (2015 to 2018)and the forecast shows a decline in pattern of rainfall series in the six states in the short run. Delayed rainfall in some partand early cessation of rainfall in some other parts expected. This corroborates the seasonal rainfall prediction (SRF) made by Nigerian MeteorologicalAgency(NIMET) for 2015.

  Key words: Time Series, Forecasting, SARIMA,Rainfall, Nigeria


 

Close
Loading...