2.4 Principal Component Analysis


Origin has a built-in Principal Component Analysis tool which is used to explain the variance-covariance structure of a set of variables through linear combinations.

And, we also provide an enhanced version of Principal Component Analysis tool, Principal Component Analysis app. This version offers the following additional features:

  • Group support in Score Plot and Biplot.
  • Confidence ellipse in Score Plot and Biplot.
  • 3D plot support for Loading Plot, Score Plot and Biplot.
  • Outlier detection in Score Plot and Biplot.


  1. With a worksheet window activated, click the menu Data > Connect to File: Text/CSV to import the sample file <Origin program folder>\Samples\Statistics\Fisher's Iris Data.dat.
  2. Click the Principal Component Analysis icon in the Apps Gallery window to open the dialog.
  3. In the Input tab, choose col(A)~col(D) in the worksheet for Input Data, where each column represents a variable. Choose Col(E) as Group to divide observations in Score Plot and Biplot. Note: You can also choose a column for Observations, which can be used for labels in Score Plot and Biplot.
    Principal Component Analysis 01.png
  4. In the Settings tab, set Analyze to Correlation Matrix; set Number of Components to Extract to 3. Note: Standardize Scores option will standardize scores of each component to set the variance to be equal to 1. Here, we don't need check it.
    Principal Component Analysis 02.png
  5. In Quantities to Compute tab, check Loadings and Scores to output them in Report Data sheet.
    Principal Component Analysis 03.png
  6. In Plots tab, specify whether to create Scree Plot, Loading Plot, Score Plot and Biplot. All except Scree Plot support 2D and 3D. The last two also support confidence ellipse and labeling of outliers.
    Principal Component Analysis 04.png
  7. Click OK button. A report sheet, a report data sheet and a plot data sheet will be created. If Show Confidence Ellipse option is checked in Plots tab, a Matrix book will also be created.
    Principal Component Analysis 05.png