ARIMA Model and its Application to Budget Performance

Abstract

In Time Series Analysis, the Autoregressive Integrated Moving Average model, has been applied to model different phenomena. Financial budget performance and classification of functions of government performance have not been modeled simultaneously in literature. In this study, a class of statistical time series models was examined in analyzing and forecasting the budget performance of Lagos State for the period under study (1968–2018). Also, it examined the trend of the Classification of Functions of Government (COFOG) in Lagos State (2008–2018). Box-Jenkins’ ARIMA and ARMA models’ approaches were used in this study. The financial budget performance was differenced to obtain stationarity. The best model chosen based on the selection criteria, Akaike Information Criterion (AIC) is ARIMA (1,1,1). The series presented for the classification of functions of government was stationary over time and the pattern indicated the volatility nature of their performance.  A unit root test was conducted to ascertain the degree of stationarity. The forecast results suggested that the financial budget would continuously increase at the rate of approximately 3% on the average yearly for next ten years; and an increment of approximately 0.24% within the classification of functions of government for the next ten years.