Modelling Interest Rate Return in Nigeria: The Asymmetric Power Autoregressive Conditional Heteroscedasticity
Tuaneh, G. L. *
Department Agric and Applied Economics, Rivers State University, Port- Harcourt, River State, Nigeria.
Ntul, A. E.
Department Statistics, University of Cross Rivers State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This study modeled the Nigerian commercial bank interest rates using Generalized Autoregressive Conditional Heteroscedasticity (GARCH). The data spanning January 1997 to December 2023 was sourced from the Central Bank of Nigeria (CBN) statistical bulletin. The time plot of the original and return series initially followed a trend and metamorphosed, which caused it to become stationary. The variables were stationary at lag one, according to the results of the series' Augmented Dickey-Fuller test. To simulate the interest rate return series, the study used both symmetric and asymmetric (GARCH) models, which captures the characteristics of financial time-series data such as volatility clustering and leverage impact. Asymmetric power autoregressive conditional heteroscedasticity (APARCH (1,1)), however, was chosen as the best model for the return series after six models were estimated for the conditional variance. Ultimately, the asymmetric APARCH model was the most suitable model to estimate the interest rate volatility. The model's post-estimation revealed that its distribution was normal.
Keywords: Interest rate, APARCH, GARCH, volatility, leverage effect