The electricity consumption of buildings composes a major part of the city's energy consumption. Electricity consumption forecasting enables the development of home energy management systems resulting in the future design of more sustainable houses and a decrease in total energy consumption. Energy performance in buildings is influenced by many factors like ambient temperature, humidity, and a variety of electrical devices. Therefore, multivariate prediction methods are preferred rather than univariate. The Honda Smart Home US data set was selected to compare three methods for minimizing forecasting errors, MAE and RMSE: Artificial Neural Networks, Support Vector Regression, and Fuzzy Rule-Based Systems for Regression by constructing many models for each method on a multivariate data set in different time terms. The comparison shows that SVR is a superior method over the alternatives.