The rapid economic improvement in Malaysia has resulted in an increase of housing prices. This trend has become a serious concern for people who want to buy houses, making many of them want to know how prices will shape up in the future in order to help them in the decision-making process and determine the appropriate policies. In this paper, the Artificial Neural Network (ANN) was studied to improve the reliability of prediction information. The Firefly Algorithm (FA) was used to optimize and improve the structure of ANN, resulting in a hybrid called FANN. The Gross Domestic Product (GDP), Population Rate (PR) and Inflation Rate (CPI) were chosen as independent variables that have a strong effect on Malaysian Housing Price Index (MHPI). The MHPI of four types of houses (Terraced, High-Rise, Detached and Semi-Detached houses) was chosen as the dependent variable. Yearly time-series datasets from 2003 to 2014 were used to train and test the proposed model. The results of FANN were compared to GANN (ANN optimization with genetic algorithm) using the same datasets. The FANN model presented a lower mean squared error (MSE) rate in tearing, validation and testing processes in the majority of the datasets of MHPI. This proves that FANN has an effective ability to be applied in housing price prediction.