Written for undergraduate and graduate courses, this text takes a step-by-step approach and assumes students have only a basic knowledge of linear algebra and statistics. The explanations therefore avoid complex mathematics and excessive abstract theory, and even statistical information is accompanied by clear numerical examples and equations are explained all the way through the process. Topics include smoothing out data with an equispaced predictor, nonparametric regression for a one-dimensional predictor, multidimensional smoothing, nonparametric regression with predictors represented as distributions, smoothing of histograms and nonparametric probability density functions and pattern recognition. Each chapter includes exercises.