Modelling Birth Rate in Otuoke, Bayelsa State, Nigeria: A Time Series Approach

Author:Eli, I.C., Barinaadaa, J.N.

published date:2024-Mar-03

FULL TEXT in - | page 1- 6

Abstract

This study seeks to determine a suitable Time series model for predicting birthrate in Bayelsa State, Nigeria, using Time series Box- Jenkins method. Monthly birth data from April, 2015 to April 2022 collected from Federal Medical Centre Otueke Bayelsa State was used. The Augmented Dick Fuller test was used to test for unit root. The unit root test revealed that the data was not stationary. The series was difference and stationary at the first differencing. Based on of various selection criteria Autoregressive Integrated Moving Average, ARIMA (1, 1, 1) was identified as an appropriate model for the forecast of birthrate. The Ljung-Box Q test used to test the adequacy of the model. The residual of the model was checked using Ljung-Box residual test and it shows no sign of autocorrelation in the residual. Thus, Autoregressive Integrated Moving Average ARIMA (1,1,1) model was recommended for forecast of birth-rate in Bayelsa state.

Keywords: Birth rate, ARIMA, Differencing, Estimation,

References