In recent years there has been a massive deployment of wind energy power systems. Wind has acquired an important role in several electric markets, and its influence on market agent decisions and price fluctuations has progressively increased, becoming one of the main factors of price variability in several areas. The penetration of wind energy in electricity markets supposes a great challenge for market agents, who have to adapt their operations to wind variability. In this sense, understanding and being able to replicate wind behavior in different markets in the medium and long term (from one month to several years) is of great importance for risk management and to develop efficient market strategies. This paper develops a novel and efficient methodology to replicate wind generation in the medium and long term, in an hourly basis, presenting probabilistic results among different areas. The objective is to represent short term wind variability, but also long-term seasonality at multi  ple scales and correlation of wind production among different areas. The proposed model uses advanced statistical methods and independent time series models for each month and area under analysis.