Stephen - If the four columns shown below are in this order in a data frame named 'data', then use
covariances <- by(data[ ,-1], data$Class, cov) to get the covariance matrices within each of the four classes. Alternative functions would be tapply() or aggregate(), but the syntax for by() is easiest to understand. - tom blackwell - u michigan medical school - ann arbor - On Mon, 1 Dec 2003, Stephen Opiyo wrote: > Dear ladies and gentlemen, > > I would like to calculate autocovarinace and cross-covariance scores 1, > 2 and 3 of four classes A, B, C and D. I am using acf and ccf from time > sires library. My problem is that I can not separate my data among the > classes A, B, C and D. When I calculated acf for Score 1, I got a wrong > result. The reason being that instead of using ony 60, 40 and 20, the > program use all the data in column under Score 1. What should I do to > calculate acf and ccf scores for each class A, B, C and D according to > he data below? > > Class Score 1 Score 2 Score 3 > A 60 11 21 > A 40 21 16 > A 20 16 18 > B 10 23 62 > B 16 8 13 > B 14 13 18 > C 22 15 22 > C 24 5 18 > C 24 12 12 > D 16 6 16 > D 12 3 8 > D 15 2 13 > > Thanks for your help. > > SO > > ______________________________________________ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help > ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
