search for: alex_s_42

I was wondering, whether there is a way to have statistical significance test for cluster agreement. I know that I can use classAgreement() function to get Rand index, which will give me some indication whether the clusters agree or not, but it would be interesting to have a formal test. Thanks.

I have some data in the Linux version of S-Plus, which I can not use anymore. The program is just broken and won't run. I'm trying to find a way to import that data to either Windows version of S-Plus (which I have running on my other machine) or R (Linux or Windows, it doesn't matter). Unfortunately, nothing seems to work. Windows S-Plus seem to ignore files from Linux .Data

It is unclear to me how aov() handles non-categorical variables. I mean it works and produces results that I would expect, but I was under impression that ANOVA is only defined for categorical variables. In addition, help(aov) says that it "call to 'lm' for each stratum", which I presume means that it calls to lm() for every group of the categorical variable, however I

Thanks for all the help on my previous questions. One more (hopefully last one) : I've been very surprised when I tried to fit a model (using aov()) for a sample of size 200 and 10 variables and their interactions. It turns out that even 2GB of RAM is not anough for aov() with this sample size, which does not seem so big for me. Am I doing something wrong or this is considered a normal

I'm having the most weird problem with bagging function. For some unknown reason it does not improve the classification (compared to rpart), but instead gives much worse results ! Running rpart on my data gives error rate of about 0.3 and bagging, instead of improving this results, gives error rate of 0.9 !!! I'm running both rpart and bagging with exactly the same parameters, I even

I have a simple linear model (fitted with lm()) with 2 independant variables : one categorical and one integer. When I run summary.lm() on this model, I get a standard linear regression summary (in which one categorical variable has to be converted into many indicator variables) which looks like : Estimate Std. Error t value Pr(>|t|) (Intercept) -3595.3 2767.1 -1.299