Regression models with long memory errors have been discussed by several authors. All proposed methods apply strictly to complete regression data sets. However, often in practice the available data are incomplete with missing values in the response or in the explanatory variables. This situation can be found in many scientific areas, such as physics, economics, etc. In this thesis, we provide a statistical methodology to handle regression data exhibiting long memory errors and missing values. This type of data appear very often in many areas, including hydrology and environmental sciences, among others. A general linear model is considered to deal with this problem and an estimation strategy is developed by ltr"
