Time Series Research Design

Written by Scott Martin
Bookmark and Share

Time series research design is another arm of mathematical statistics that statisticians can use to aid the development of your research studies. Some fields which can benefit from time series research design are those in the business and finance realm, such as economic forecasting, stock market analysis, and yield projects. Of course, labor industries also benefit from time series research design, in the form of workload projections, longitudinal studies, or utility studies.

Though time series research design has many applications, there are two central goals of any such study. First, identification of the nature of a particular relation or phenomenon must occur, by analyzing occurrences of observations as sequences. Secondly, forecasting can be accomplished, by examining these patterns to predict future time series variable values.

Of course, before either goal is accomplished, the observed time series data pattern must be first identified and described in detail. Interpretation and integration of data can then subsequently occur. Independent of theory, future predictions can be made from such data analysis.

A Statistician's Role in Time Series Research Design

As in most design forms, a statistician can be an indispensable asset in choosing the most appropriate technique to implement. Furthermore, statisticians design data collection and analysis procedures, and create models and their testing specifications. Examples of such model and specification creation by statisticians include canonical correlation analysis and selection criteria.


Bookmark and Share