AUDE MARIE PLONTZ
2007-Nov-14 23:20 UTC
[R] Question about AGNES by Rousseeuw et al. in the package "cluster": How many clusters?
Dear all, I am no stat wiz and I am just trying to use the AGNES algorithm at my very modest level of statistical of understanding. I have difficulties understanding the ouput from AGNES. My question is: how to interpret the output, especially how do you I know which cluster solution is the best? In SPSS, an Agglomeration Schedule table is produced and I used to look at the biggest jump between the error coefficients for each agglomerative steps in order to find where to stop clustering. But with the Agnes output I don't know what I should be looking at. Thanks so much for your help, Aude Aude Plontz Refugee Health Research Centre, Melbourne, Australia Phone: at home: +61 (0)3 9917 2134 at La Trobe University, Bundoora campus (every other Thursday): +61 (0)3 9479 3161 (desk, Martin Building, room 475, consultation hours: 9.00am-11.00am, 1.00pm-5.00pm) or +61 (0)3 9479 5874 (Estelle Purchas, RHRC secretary) at the Victorian Foundation for Survivors of Torture, Brunswick (on some Wednesdays): +61 (0)3 9940 1567 (desk, Brunswick Business Incubator) or +61 (0)3 9388 0022 (Brunswick VFST reception) Email: amplontz at students.latrobe.edu.au (email account checked once daily, except when on leave) audrus at freenet.de