Statistical Methods for Intelligent Data Analysis

Statistics deals with collection, organization, and analysis of data. Organization and description of data is studied under descriptive statistics, whereas analysis of data and making predictions based on it is dealt with in inferential statistics. The basic concepts of statistics are used to develop different methods for inference. The generalized linear models include techniques such as regression, ANOVA, log‐linear models, analysis of survival data, and so on, whereas specialized methods such as multivariate analysis of variance and random effect models are also used. Not all methods can be applied to all data sets, and suitability of a model depends upon the type of data. Predictions made using statistical inference can be prone to errors, and it is important to identify the sources of error and use preventive techniques. The study of error helps minimize them in not only conventional statistical methods, but also in neural networks and other machine learning methods.In this chapter, the basic concepts of probability are first introduced. The various techniques of descriptive and inferential statistics, their types, and applications are then discussed. Various commonly used methods have been introduced along with suitable examples. Types of errors that can arise in the application of statistical methods are discussed and analyzed with examples. The chapter intends to give fellow readers a firm footing in the statistical concepts that are widely used in the field on intelligent data analysis, which can then be used to develop state‐of‐the‐art methods for specific use cases.