Data for: Energy production in Brazil: Empirical facts based on persistence, seasonality and breaks

Abstract of associated article: This paper investigates the statistical properties of the production of energy in Brazil using long range dependence techniques in monthly data from January 2000 to February 2013. Two important features of the data are analyzed, in particular, its degree of persistence and seasonality. The results indicate first that seasonality is an essential issue in modeling the persistence in energy production. Also, the persistence itself, measured in terms of the differencing parameter is relevant, with orders of integration in the series found to be positive though smaller than 1 and thus implying mean reversion. A single structural break is also found in two of the series. Policy implications of the results obtained are also derived.

相关论文摘要：本文基于2000年1月至2013年2月的月度电力生产数据，运用长程依赖（long range dependence）分析方法，探究巴西电力生产的统计特性。本研究重点考察了该数据集的两项重要特征：持续性程度与季节性。研究结果首先表明，在对电力生产的持续性进行建模时，季节性是不可或缺的关键要素。此外，以差分参数（differencing parameter）衡量的持续性本身同样具有研究价值：序列的积分阶数为正值且小于1，这意味着序列具备均值回复（mean reversion）特性。同时，两条序列中均存在单个结构断点（structural break）。本研究还推导了所得结论的政策启示。