## With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

No credit card required

# R CODE FOR MULTIVARIATE ADAPTIVE TESTS

Note: The functions cdfhat must also be included. This is in Appendix A and on the file adapt.r.

```adaptmult <- function(y,complete,reduced,r=2000,seed=1492) {
lambdau<- teststat(y,complete,reduced)
e <- 0
redu <- lm(as.formula(reduced))
yhat <- predict(redu)
yresidual <- residuals(redu)
set.seed(seed)
n <- length(yhat[,1])
nvars <- length(yhat[1,])
y <- matrix(0,n,nvars)
observations <- c(1:n)
for (k in 1:r) {
place <- sample(observations)
for (i in 1:n) {```
`row <- place[i] y[i,] <- yhat[i,] + yresidual[row,] } lambdap <- teststat(y,complete,reduced) if( (lambdap <= lambdau) ) e <- e + 1 } p <- (e+1)/(r+1) detach(testdata) cat(“\n”,“***************************************”,“\n”,“\n”) cat(“ Input:”,“\n”,“\n”) cat(“ Data set:”,dataset,“\n”,“\n”) cat(“ Dependent variables: ”, yvars,“\n”,“\n”) cat(“ Complete model: ”,complete,“\n”,“\n”) cat(“ Reduced model : ”,reduced,“\n”,“\n”) cat(“ Number of permutations = ”,r, “\n”,“\n”) cat(“ Random seed = ”,seed,“\n”,“\n”,“\n”,“\n”) cat(“ Output from Adaptive Test:”,“\n”,“\n”) cat(“ Unpermuted lambda = ”,round (lambdau, 4),“\n”,“\n” ) cat(“ p-value = ”,round(p,4),“\n”,“\n”,“\n”) cat(“\n”,“*************************************”,“\n” “\n”) return(p) } teststat <- function(y,complete,reduced) { red <- lm(as.formula(reduced)) resid <- residuals(red) n <- length(resid[,1 ] ) nvars <- length(resid[1,]) # standardize residuals temp <- matrix(0,n,1) for (j in 1:nvars) { temp <- resid[,j] q25 <- quantile(temp,0.25,type=6) ...`

## With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

No credit card required