Treatments that are accessible and up-to-date showing the relationship between neural networks and statistical Statistical Approach to Neural Networks for Pattern Recognition presents the statistical treatment of the Multilayer Perceptron (MLP), the most widely used in neural network models. This book aims to answer the questions that arise when the statistic was first confronted with the type of model, such as: How strong was the model for outliers? further treatment of a very important principal aspects of the MLP are provided, such as the robustness of the model in terms of isolated or atypical data, the influence and sensitivity curves MLP, MLP is why the model is strong enough, and modifications to make the MLP is more robust. Throughout this book, the MLP model is extended in several directions to show that the statistical modeling approach can make a valuable contribution, and further exploration for the MLP model fitting is made possible through the R and S-PLUS code is available in book-related Web sites.