search

kaskr / adcomp

AD computation with Template Model Builder (TMB)
Other
176 stars 80 forks source link

Improvement of help for oneStepPredict #364

Closed vtrijoulet closed 2 years ago

vtrijoulet commented 2 years ago

I think the help page of the function oneStepPredict function could do with more details.

Some suggestions below:

##' @param range Possible range of the observations (only and recommended for \code{method="oneStepGeneric"}).
##' @param seed Randomization seed (discrete case only). If \code{NULL} the RNG seed is untouched by this routine (recommended for simulation studies).
##' @param trace If \code{TRUE} progress printed to console, inversely if \code{FALSE}. If \code{== 2} the one-step ahead predictive distribution on log-scale is plotted and if \code{>= 3} the one-step ahead predictive distribution on natural scale is plotted (only available for \code{method="oneStepGeneric"})

Possibility to add the suggestions to “Details” instead if preferred.

timjmiller commented 2 years ago

Also would be great to have more detailed description on what is in each column (other than "residual") of the returned data.frame under the alternative method options.