I have a very big data and I would like to perform multi comparison test. However, I must firstly perform ANOVA to be able to calculate p-value etc

There are a lot of posts but none of them applicable for my question.

Lets say I have a matrix as follow:

y <- outer(1:20000,1:1500)

How can I perform this simple statistical analysis on it in R ?

  • $\begingroup$ Do you really have a sample size of 20000? $\endgroup$ Feb 17 '15 at 18:51
  • $\begingroup$ Yes, if it was small, I would not ask the question here ! $\endgroup$
    – Learner
    Feb 17 '15 at 19:40
  • $\begingroup$ I meant in the converse: 1500 is not the sample size but the number of measurements per subject. The point is that if you have more dependent measurements than independent replications, you have high dimensional data requiring special treatment. Otherwise, N Brouwer's answer is the way to go. $\endgroup$ Feb 21 '15 at 1:24

biglm or speedglm might be what you need. Multiple comparisons might be another issue but multcomp is pretty powerful.


Not the answer you're looking for? Browse other questions tagged or ask your own question.