I am performing classical MDS on a dataset (Gower matrix returned by R cluster package function daisy). In my field, a measure of fit of the MDS is reported. Other researchers usually perform this analysis with the Clustan software (Wishart 2004) software, which performs the Kruskal's Stress test. Kruskal's stress is defined as:
$$ \sqrt{\frac{\sum (d_{ij}-\delta_{ij})^{2}}{\sum d_{ij}^{2}}} $$
where πππ represents the distances, and πΏππ represents the disparities.
Despite several lost hours googling around, I can't get my hands on this Clustan software, so I'm trying to recreate this analysis in R. I can perform MDS fine with cmdscale in R, but I don't understand the 'Goodness of Fit' ($GOF$) measure returned by cmdscale and would prefer to use what's standard in my field. Are there any packages that include Kruskal's Stress for MDS in R? Is there a way to calculate it manually?
