Buildings represent a large percentage of a country’s energy consumption and associated greenhouse gas emissions. The energy needed in order to maintain internal conditions within buildings is responsible for a significant portion of the overall energy usage and greenhouse emissions. Thus, improving energy efficiency in buildings is of great importance to our overall sustainability. Over the past few decades, a lot of research has been carried out in order to improve building energy efficiency through various techniques and strategies. The forecasting of energy usage in an existing building is essential for a variety of applications like demand response, fault detection & diagnosis, optimization and energy management. This is a typical time-series based application. Data-driven forecasting models typically include two main approaches; statistical and machine learning based schemes. The statistical approach typically applies a pre-defined mathematical function and has shown good performance for medium to long-term energy forecasting. In addition, such models have shown acceptable performance for short-term forecasting of consumption electricity loads. The machine learning approach in contrast, typically applies an algorithmic approach (which may non-linearly transform the data), in order to provide a forecast