-
Statistics: Type I ~ IV sum of squares
Model: Y ~ 1 + A + B + C + A*B + A*C + B*C + A*B*C Type I - A partitioning of the model sum of squares into component sums of squares due t...
-
Statistics: Friedman Test for Randomized Block Designs
block <- rep(1:6,3); b <- length(unique(block)) task <- rep(c('A','B','C'),rep(6,3)); k<- length(unique(task)) time <- c(1.21,1.63,1.42,1.16...
-
Statistics: cross over design (binary output)
data crossover; input group $ seq1 seq2 count; seq='(' || put(seq1,1.) || ',' || put(seq2,1.) || ')'; diff=seq2-seq1; trt=diff*(grou...
-
Statistics: Wilcoxon Signed-Rank Test
### H0: median = tau vs Ha: median > tau ### Normal approximation tau<-0.015 N <-20 x <- rnorm(N, 0.05, 0.1) V <- sum(rank(abs(x-tau))[x-tau...
-
Statistics: Modeling McNemar's test
proc import datafile='F:\consulting\dich.csv' out=dich dbms=csv replace; run; /* McNemar test */ proc freq data=dich; by group; table be...
-
Statistics: Kurskal-Wallis Test for the One-Way Layout
line <- rep(1:3,rep(10,3)) defects <- c(6,38,3,17,11,30,15,16,25,5,34,28,42,13,40,31,9,32,39,27,13,35,19,4,29,0,7,33,18,24) R <- rank(defect...