AJIBODE, I. A and Ogunnusi, O.N (2019) ON MODELLING RAINFALL PATTERN IN TWO DIFFERENT LOCATIONSWITH SAME SEASONAL ARIMA Models. In: Book of Proceedings of 4th National Development Conference of The School of Pure and Applied Science, The Federal Polytechnic Ilaro, Ogun State, 2nd - 5th December, 2019, The Federal Polytechnic, Ilaro.
![[img]](https://eprints.federalpolyilaro.edu.ng/style/images/fileicons/text.png) |
Text
AJIBODE, I.A. CONFERENCE PAPER 4.pdf Download (354kB) |
Abstract
The Forecast of rainfall on monthly and seasonal timeframe is not only scientifically challenging, but is also a means of planning for decision making purposes. Various research groups have attempted to forecast rainfall on a seasonal timeframe using different techniques. This research work describes the Box-Jenkins time series Seasonal ARIMA (Auto Regressive Integrated Moving Average) approach for the prediction of rainfall on monthly scale using Abeokuta and Ijebu Ode, Ogun State. Seasonal ARIMA(1,0,2)(1,0,1)12 was identified the best model to forecast rainfall for Abeokuta and Ijebu Ode for the next four years with 95% confidence level by analyzing 17 years data(1999-2016) recorded on monthly basis by the Nigeria Meteorological Agency as extracted from Central Bank of Nigeria statistical bulletin. Previous data was used to formulate the seasonal ARIMA model and in the determination of model parameters. The preference evaluation of the adopted models in R was carried out on the basis of Akaike Information Criterion (AIC), MSE and Log-likelihood taking method of maximum likelihood estimation into consideration. The result indicated that the Seasonal ARIMA model provide consistent and satisfactory prediction for rainfall parameters on monthly scale as evidenced from the Shapiro-wilk normality test and Ljung-Box test of independence of residuals. Forecasts also indicated that there is higher rise of rainfall in the month of April, May, June, July and September in Abeokuta and Ijebu-Ode respectively between year 2017 to 2020, but with little variation of measurements.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Faculty of Engineering, Science and Mathematics > School of Mathematics |
| Depositing User: | Mr. Bolanle Yisau I. |
| Date Deposited: | 29 May 2021 16:36 |
| Last Modified: | 29 May 2021 16:36 |
| URI: | http://eprints.federalpolyilaro.edu.ng/id/eprint/1493 |
Actions (login required)
 |
View Item |
