Question

A researcher wants to develop a way of using morphometric measurements farm dogs to accurately estimate their fat free mass (FFM) to skeletal size. Considering that it is straight forward to measure a dog's head, length etc, but very expensive to gather data on a dog's FFM, what multivariate method would be suitable?

Answer 1

In this case, the researcher can use a Principal Component Analysis (PCA). This analysis is a dimensionality-reduction method which is commonly used to reduce the dimensionality of large data sets. This method transforms a large set of variables( for example length of different parts of the dog skeleton) into a smaller set of variables which still contains most of the information in the large set. Reducing the number of variables could affect the accuracy of the data but smaller data sets are easier to analyze, explore and visualize, making the analysis easier and faster.