Often you will find that the data you collected has a time component. You might collect information about a crime rate over 25 years, but if you include that variable as an independent variable in your regression model, then you're most likely violating an assumption of regression (nonautocorrelation of error terms). Time series methods will allow you to use those variables that have time components.
A keyword search for time series analysis in DUQSearch will return many books and eBooks.
An Introduction to Time Series Analysis and Forecasting by 
Time Series: Modeling, Computation, and Inference by 
