Syad Hamina

Publications

• Review Article   
  Data Reduction Using Principal Component Analysis: Theoretical Underpinnings and Practical Applications in Public Health
  Author(s): Syad Hamina*
  
  Big datasets are becoming increasingly common and can be challenging to understand and apply in public health. One method for lowering the dimensionality of these datasets and improving interpretability while minimizing information loss is data reduction using Principal Component Analysis (PCA). It achieves this by successively maximizing variance through the creation of new, uncorrelated variables. PCA is an adaptive data analysis technique because it simplifies the process of finding new variables, or principal components, by solving an eigenvalue or eigenvector problem. These new variables are determined by the dataset being used, rather than by the analyst starting from scratch. It is also adaptable in another way because varieties of the method have been designed to adjust to various data structures and types. However, there are serious problems in t.. Read More»
  

  Abstract HTML PDF