This chapter explores the temporal behavior of time series, focusing on statistical and mathematical techniques used to analyze dependencies and structures within time-dependent data. The chapter begins with autocorrelation, a fundamental concept for measuring how past values in a time series influence future values, introduced as a key tool for identifying patterns and dependencies. Next, power spectrum analysis is discussed, which helps decompose a time series into its frequency components, revealing periodicities. This technique is widely used in signal processing and finance. The concept of mutual information follows, offering a nonlinear method to assess dependencies between time series variables. Mutual information captures complex relationships and is useful for analyzing nonlinearly dependent data. The Hurst exponent is then introduced as a measure of long-term memory and persistence in time series data. It helps differentiate between random, persistent, and anti-persistent behaviors, which are important in fields such as financial market analysis and hydrology. The chapter also covers Hjorth parameters, which provide a statistical measure of signal complexity, mobility, and activity. These parameters are often used in biomedical signal analysis, particularly in EEG studies. Finally, the chapter discusses clustering techniques for time series data. Several clustering methods are presented, including: By the end of this chapter, readers will have a strong understanding of the fundamental techniques used to analyze the temporal properties of time series, helping them extract meaningful insights and patterns from complex datasets.