A stochastic model of wind velocity in Lake Maracaibo

• Salvador Pinto Universidad del Zulia-Venezuela

Abstract

In this paper hourly records of wind velocity were analyzed as Multivariate Time Series. Data were collected from three meteorological stations on the coast of Lake Maracaibo. The fundamental result was the identification with a multivariate ARMA model that was estimated and validated for every month of the year. Before the identification, the seasonal component was eliminated by means of a suitable transformation. The result of this investigation is useful for contingency plans in the area.