Corrective actions of well logging data in the presence of serial correlation

The sonic profile is used, mainly in pioneering wells, which receive greater investments for acquisition from data, therefore these wells serves as reference to analyze the petroliferous potential of the area. for considered wells of exploitation, normally the sonic tools for economic reasons are not races, mainly in the times of low price of the oil. as it is not viable economically to make an acquisition of the sonic profile, in the great majority of the old wells, mainly in result of the stop of the production of the oil, it must be appealed to other ways to solve the problem. one in the ways is to use statistical methods that generate a sonic profile, said synthetic, to establish estimates of the seismic speeds of the rocks to the long one of the wells. Thus, the objective of this article is to present measured corrective for models with series with serial correlation, the data used in this study is the profiles of wells of oil of the basin sedimentary Sergipe-Alagoas (Brasil). The information acquired in this study will be excellent for the construction of statistical models, that can establish estimates of the seismic speeds of the rocks throughout the wells. The well of oil used possesss different stratigrafics levels. The result showed that the choice of the iterative method of Cochrane-Orcutt, for the esteem for squared minimums generalized was efficient for removing the serial correlation, what it was not possible for the esteem for squared minimums ordinary. This article is supported by CNPq an modeled by SAS.