risk: Problem when models are decreased and increased afterwards

[hofnerb]

## Generate some data
set.seed(1907)
x1 <- rnorm(1000)
x2 <- rnorm(1000)
x3 <- rnorm(1000)
x4 <- rnorm(1000)
x5 <- rnorm(1000)
x6 <- rnorm(1000)
mu    <- exp(1.5 +1 * x1 +0.5 * x2 -0.5 * x3 -1 * x4)
sigma <- exp(-0.4 * x3 -0.2 * x4 +0.2 * x5 +0.4 * x6)
y <- numeric(1000)
for( i in 1:1000)
    y[i] <- rnbinom(1, size = sigma[i], mu = mu[i])
dat <- data.frame(x1, x2, x3, x4, x5, x6, y)

## Fit model
model <- glmboostLSS(y ~ ., families = NBinomialLSS(), data = dat,
                      control = boost_control(mstop = list(mu = 10, sigma = 20), trace = TRUE),
                      center = TRUE)
risk(model)  ## OK
lapply(risk(model), length)  ## OK

## Increase mstop
mstop(model) <- c(20, 30)
risk(model) ## Missing value introduced
lapply(risk(model), length)  ## in principle OK


## No issue when model does not need to be reduced (above to mstop = c(10, 10))
model <- glmboostLSS(y ~ ., families = NBinomialLSS(), data = dat,
                      control = boost_control(mstop = 10, trace = TRUE),
                      center = TRUE)
mstop(model) <- 20
risk(model)

@ja-thomas The above error exists at least in the current implementation in the master branch. Please also check this in the devel branch (or your new branch). Altogether, we need to check all occurences of "risk" and make sure that it works with the new mboost risk. I think it's best if we only do this in the devel branch as the old master branch most likely will not make it to CRAN another time...

[ja-thomas]

This is already fixed in my branch and works both with the cylical and the noncyclical version. Can be closed as soon as the branch in merged.

[ja-thomas]

Fixed