Finding significance levels for cointegrating coefficients in cajorls I am investigating the long-term relationship of some variables using the R package vars, but in the output of the cajorls function I cannot see whether each coefficient is significant.  This information is provided by the cajools function but for the cointegration equations I need to use the cajorls function. Does anyone have an idea how to get the information from cajorls?
 A: I am not aware of any function or slot within functions in "vars" that allow you to retrieve directly standard errors or p-values for the beta coefficients. 
So, in case there is no such thing, you can still use the blrtest function that provides a LR test for the beta coefficients. You could test manually:
library(vars)
data(finland)
sjf.vecm <- ca.jo(finland, ecdet = "none", type = "eigen", K = 2)

# extract B
B_0 <- cajorls(sjf.vecm, r=1)$beta
# Set one element to zero
B_0[2] <- 0
#test
blrtest(sjf.vecm, H=B_0, r=1)@pval[1]

You could use the same strategy to obtain confidence intervals by "profiling" (as is done for glm, see profile.glm) the LR test:
B_0 <- cajorls(sjf.vecm, r=1)$beta
    foo_pval <- function(x) blrtest(sjf.vecm, H=B_0+c(0,x,0,0), r=1)@pval[1]-0.05
    uniroot(foo_pval, interval=c(0,1))$root

And don't forget not to test for the normalised values, set to one for identification!!
A: To my knowledge, the answer is:    
The followings are deg. of freedom non-adjusted:
coeftest(cajorls(ca.jo(finland, ecdet="none", type="eigen", K=2),r=1)$rlm) 
coef(summary(cajorls(ca.jo(finland, ecdet="none", type="eigen", K=2), r=1)$rlm))

The followings are deg. of freedom adjusted:
V1.eigen <- ca.jo(finland, ecdet="none", type="eigen", K=2) # rank=1
vecm <- cajorls(V1.eigen, r=1)
beta <- V1.eigen@V[,1] 
alfa <- V1.eigen@W[,1] 
residuals <- resid(vecm$rlm)
N <- nrow(residuals)
sigma <- crossprod(residuals)/N     
beta.se <- sqrt(diag(kronecker(solve(crossprod(V1.eigen@RK[,-1])), solve(t(alfa)%*%solve(sigma) %*% alfa))))
beta.t <- c(NA, beta[-1]/beta.se) # deg.of freedom adjusted
names(beta.t) <- rownames(cajorls(ca.jo(finland, ecdet="none", type="eigen", K=2), r=1)$beta) 
beta.t
beta.pval <- dt(beta.t, df=vecm$rlm$df.residual)
beta.pval

Inspiring from here, one may suggest (for a neat presentation):
library(texreg)
cajo_beta_create <- function(cajo_o, cajorls_o) {
alfa <- coef(cajorls_o$rlm)[1, ]
residuals <- resid(cajorls_o$rlm)
N <- nrow(residuals)
sigma <- crossprod(residuals) / N
beta <- cajorls_o$beta
# standard errors
beta.se <- sqrt(diag(kronecker(solve(crossprod(cajo_o@RK[, -1])), solve(t(alfa) %*% solve(sigma) %*% alfa))))
beta.se2 <- c(NA, beta.se)
beta.t <- c(NA, beta[-1] / beta.se)
beta.pvalue <- dt(beta.t, df=cajorls_o$rlm$df.residual)     # p values

tr <- createTexreg(coef.names = as.character(rownames(beta)), coef = as.numeric(beta), se = beta.se2, pvalues=beta.pvalue,
gof.names = c('Dummy'), gof=c(1), gof.decimal=c(FALSE))
return(tr)
 }
cajo_beta_create(V1.eigen, vecm)
screenreg(cajo_beta_create(V1.eigen, vecm))

