Premium
Time Series Analysis in Historiometry: A Comment on Simonton
Author(s) -
Velicer Wayne F.,
Plummer Brett A.
Publication year - 1998
Publication title -
journal of personality
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.082
H-Index - 144
eISSN - 1467-6494
pISSN - 0022-3506
DOI - 10.1111/1467-6494.00020
Subject(s) - generalization , strengths and weaknesses , series (stratigraphy) , causation , time series , psychology , multivariate statistics , multivariate analysis , computer science , social psychology , mathematics , machine learning , mathematical analysis , paleontology , political science , law , biology
Time series analysis (TSA) is one of a number of new methods of data analysis appropriate for longitudinal data. Simonton (1998) applied TSA to an analysis of the causal relationship between two types of stress and both the physical and mental health of George III. This innovative application demonstrates both the strengths and weaknesses of time series analysis. Time series is applicable to a unique class of problems, can use information about temporal ordering to make statements about causation, and focuses on patterns of change over time, all strengths of the Simonton study. Time series analysis also suffers from a number of weaknesses, including problems with generalization from a single study, difficulty in obtaining appropriate measures, and problems with accurately identifying the correct model to represent the data. While careful attempts are made to minimize these problems, each is present in the Simonton study, although sometimes in a subtle manner. Changes in how the data could be gathered are suggested that might help to solve some of these problems in future studies. Finally, the advantages and disadvantages of employing alternative methods for analyzing multivariate time series data, including dynamic factor analysis, are discussed.