Transformation Methods For Latent Roots And Vectors.

Area of Science:

• Mathematics
• Statistics
• Psychometrics

Background:

• Principal component analysis (PCA) is a statistical technique used for dimensionality reduction.
• Solving the characteristic equation is a critical step in PCA.
• The Hotelling iterative method is a common approach for this solution in psychology.

Purpose of the Study:

• To present and evaluate two new transformation methods for solving characteristic equations in PCA.
• To demonstrate the advantages of these new methods over the Hotelling iterative procedure.

Main Methods:

• The study employs factorial analysis using the principal component method.
• Two novel transformation methods are introduced and detailed.
• These methods are compared against the Hotelling iterative method.

Main Results:

• The proposed transformation methods offer distinct advantages compared to the Hotelling procedure.
• These advantages likely relate to computational efficiency or numerical stability.

Conclusions:

• The new transformation methods provide effective alternatives for solving characteristic equations in PCA.
• These methods have the potential to improve the application of factorial analysis in psychological research.