Asian Journal of Probability and Statistics

This paper investigated the model performances and the features of monthly demand deposits in Nigeria during the period of economic diversification from the year 2016 to 2022 and the data set used in the study were originally collected from the Central Bank of Nigeria (CBN) database. However, the study applied the Box-Jenkins methodology and Augmented Dickey-Fuller (ADF) test statistics for stationarity. The AIC, BIC, R² and Adj.R² were considered when determining the adequacy of the models and checking the fitness of the GARCH and ARIMA models on the monthly Nigerian demand deposits. The GARCH (1, 1) and ARIMA (2, 1, 2 ) models were estimated accordingly using diagnostic tests and results were obtained. The result of the fitted plot clearly identified a linear pattern with an upward trend and non-stationarity. The result of the differenced plot also identified heteroscedasticity data with a regular pattern of variation of an error term. According to the results from the stationarity test. The first stationarity test failed because the p-value is greater than the significance level 0.5 then exhibits trend and seasonality while the p-value of the second stationarity test is less than the significance level 0.5 implying stationary, which means it has a constant mean, variance and autocovariance over the time. The estimation of GARCH (1, 1) and ARIMA (2,1,2) have a small R²andAdjR², indicating poor model fit which exposed that the models did not capture the underlying patterns in the data. From the estimation of GARCH(1,1) and ARIMA(2,1,2), the AIC and BIC values were close, meaning that they are intertwined with strong autocorrelation and volatility cluster in Nigerian Monthly Demand Deposits. The results indicate that the series is unpredictable, exhibiting characteristics of a white noise process, with data varying randomly around the mean and variance. Policy implication of this study is that the demand deposits might make it challenging for the Central Bank of Nigeria (CBN) to predict and manage liquidity in the financial system which will make them adjust the interest rate more frequently to accommodate strategies to account for unpredictability of demand deposits.