Abstract
The study forecasts interest rates amid economic and financial instabilities using data sourced from South Africa. Interest rates are a crucial financial instrument used by central banks to manage the economy. Monthly data was used spanning from 2000 – 2022 using time series regression techniques of autoregressive conditionally heteroscedastic (ARCH) to measure volatility and be able to test forecasting ability. The Generalized Autoregressive Conditional Heteroskedasticity (GARCH) technique and the Threshold GARCH were used to cater for symmetry and asymmetry
103
INTERNATIONAL JOURNAL OF ECONOMICS AND FINANCE STUDIES
Vol: 17 No: 01 Year: 2025 ISSN: 1309-8055 (Online) (pp. 102-131) Doi: 10.34109/ijefs.202517107
assumptions. Economic instabilities have a significant impact on interest rates, and failing to account for them in forecasting models can lead to inaccurate predictions. Interest rate forecasting is essential for various patrons, which include financial institutions, financial markets investors, and policymakers. Accurate interest rate forecasts can assist policymakers make good decisions regarding monetary policy, while investors can use them to make investment decisions. Instabilities were inherent within the series, as empirically shown form the results. With occurrences like the global COVID-19 crisis, widespread protests, and the 2008 global financial crises, instabilities were more pronounced around 2009, and between 2016 and 2020. Results showed that the normal error distribution supposition is the best performing hypothesis across all three series (interest rates, treasury bill rate, and government bonds rate. The results point to the need for better macroeconomic policy management to reduce instabilities and have better predictable trends to allow investors planning.