Unit 5 ip (4 pages) | CS871 Advanced Quantitative Analysis | Colorado Technical University

 

Parsimony in statistical modeling is often discussed in terms of Occam’s razor in the formulation of hypotheses. Address the following:

  • Discuss the issues of overfitting versus using parsimony and how this is particularly important in big data analysis.
  • Is overfitting more of a problem in the generalized least squares model?
  • Discuss some hierarchical methods that do not require the parsimony of generalized least squares model.
  • Discuss these hierarchical methods, and provide links to any references that you find on the topic.

Be substantive and clear, and use scholarly examples to reinforce your ideas.