The notion of data smoothing is a fundamental concept in data analysis. In essence, it consists in creating an approximating function that attempts to capture only the important patterns, while filtering noise and ignoring the data structures that are deemed not relevant. The functions commonly referred to as filters can serve as examples of typical smoothers. In our treatment of the topic, we focus on one of the most well-known nonparametric smoothing techniques called kernel density estimation (KDE). There exists a number of methods for nonparametric density estimation, based on e.g. kernel smoothing, histograms, orthogonal series, splines, frequency polygons, wavelets or the penalized likelihood. Our method of choice is kernel density estimation, given that it can be easily interpreted and is very often used in practical applications.