This article examines hedging in South African stock index futures market. The hedge ratios are estimated by six econometric techniques: the standard OLS regression, simple and vector error correction models, the ECM with generalised autoregressive heteroskedasticity (GARCH) as well as time-varying CCC-ARCH and Diag-BEKK ARCH models. The empirical results show that the ECMGARCH model (capturing volatility clustering) provides best hedging ratios, while CCC-ARCH is superior to OLS, ECM and VECM. We conclude that there is not a unique model specifi cation for measuring hedge ratios. For each market (emerging and mature), a model’s comparative analysis must be conducted in order to extract the best performing model.