With Principal Components Analysis (PCA), we can try to explain the variance-covariance structure of a set of variables. We will use PCA to investigate linear associations between a large number of variables; or rather, we will change the dimensionality of a large dataset to a reduced number of variables. This can help identify the relationships in a dataset that are not immediately apparent.
As such, PCA can be a useful exploratory tool in data analysis and can often lead to more in-depth analysis.
This example looks at the tax revenue in the UK from April 2008 to June 2013.
The following steps will generate the principal components of the input factors and also plots to evaluate the impact of the first two principal components:
Open the
Tax Revenue.MTW
worksheet.Go to the Stat menu, click on Multivariate, and select Principal Components...
For the Variables: section, select the numeric columns from
PAYE Income
toCustoms duty
.In...