This paper is the .rst one to: (i) provide in-sample estimates of linear and nonlinear Taylor rules augmented with an indicator of .nancial stability for the case of South Africa, (ii) analyse the ability of linear and nonlinear monetary policy rule speci.cations as well as nonparametric and semiparametric models in forecasting the nominal interest rate setting that describes the South African Reserve Bank (SARB) policy decisions. Our results indicate, .rst, that asset prices are taken into account when setting interest rates; second, the existence of nonlinearities in the monetary policy rule; and third, forecasts constructed from combinations of all models perform particularly well and that there are gains from semiparametric models in forecasting the interest rates as the forecasting horizon lengthens.
