12-ma`ruza Since classical statistics provides many data analysis methods and supports and jus- tifies a lot of others, we provide in this appendix a brief review of some basics of statistics. We discuss descriptive statistics, inferential statistics, and needed funda- mentals from probability theory. Since we strove to make this appendix as self- contained as possible, some overlap with the chapters of this book is unavoidable. However, material is not simply repeated here but presented from a slightly different point of view, emphasizing different aspects and using different examples.

In [14] (classical) statistics is characterized as follows:

Statistics is the art to acquire and collect data, to depict them, and to analyze and interpret them in order to gain new knowledge.
This characterization already indicates that statistics is very important for data anal- ysis. Indeed: there is a vast collection of statistical procedures with which the tasks described in Sect. 1.3 (see page 11) can be tackled or which are needed to support or justify other methods. Some of these methods are discussed in this appendix. How- ever, we do not claim to have provided a complete overview. For a more detailed review of (classical) statistics and its procedures, an interested reader is referred to standard textbooks and references like [3, 4, 10].
The statistical concepts and procedures we are going to discuss can roughly be divided into two categories corresponding to the two main areas of statistics:

descriptivestatistics (Sect. A.2) and

inferentialstatistics (Sect. A.4).

In descriptive statistics it is tried to make data more comprehensible and inter- pretable by representing them in tables, charts, and diagrams, and to summarize them by computing certain characteristic measures. In inferential statistics, on the other hand, it is tried to draw inferences about the data generating process, like es- timating the parameters of the process or selecting a model that fits it. The basis of many procedures of inferential statistics is probability theory (see Sect. A.3); its goal is usually to prepare for and to support decision making.

M.R. Berthold et al., GuidetoIntelligentDataAnalysis,