search for: varhb

...imated parameters ? thanks, peter P.S. the simple solution for this particular problem: ## get the mean for each level muV <- by(y,cl,mean) ## get the variance for each level varV <- by(y,cl,var) ## the mean of Hb is muHb <- H %*% muV ## because of independence, the variance of Hb is varHb <- sum(varV) ## the probability of error, so-called p-value: p.val <- 2 * pnorm( -abs(muHb), mean=0, sd=sqrt(varHb),lower.tail = TRUE) thanks again, peter -- Peter Salzman, PhD Department of Biostatistics and Computational Biology University of Rochester [[alternative HTML version dele...