The FTSE Bursa Malaysia KLCI index is a form of capitalized trading index that is made up of over thirty trading companies in Malaysia. These type of time series data is classified as highly chaotic due to the nature and occurrence of trend and seasonality within trading patterns, hence making the analysis and forecasting process cumbersome. The main aim of financial analysts in forecasting such data is to obtain an effective and feasible solution that will assist in future planning and expectation of trends that are most likely to occur in the future. Such analysis is vital to the choices made during the modelling phase that fits historic data within the forecasting model. This paper presents an empirical analysis of KLCI time-series using an enhanced ELMAN-NARX hybrid model by performing multi-step-ahead forecasts. The proposed hybrid model is trained using a Gauss approximated Bayesian regulation algorithm. Performance analysis based on error metrics shows that proposed hybrid model provides robust multi-step-ahead forecasts in comparison to previously used models. © 2016 IEEE.