In this chapter, we introduce several useful ideas that incorporate external information into time series modeling. We start with models that include the effects of interventions on time series’ normal behavior. We also consider models that assimilate the effects of outliers—observations, either in the observed series or in the error terms, that are highly unusual relative to normal behavior. Lastly, we develop methods to look for and deal with spurious correlation—correlation between series that is artificial and will not help model or understand the time series of interest. We will see that prewhitening of series helps us find meaningful relationships.
CITATION STYLE
Zivot, E., & Wang, J. (2003). Time Series Regression Modeling. In Modeling Financial Time Series with S-Plus® (pp. 167–207). Springer New York. https://doi.org/10.1007/978-0-387-21763-5_6
Mendeley helps you to discover research relevant for your work.