Depending on the current transformation h(y)={y,y,y2/3},
V(h(y0)−h(μ0))=V(h(y0))+V(h(μ0))
is used to compute a prediction interval. The prediction variance
consists of a component due to the variance of having a single
observation and a prediction variance.
Quantile level in Gaussian based CI, i.e. an (1−α)⋅100%
confidence interval is computed.
skewness.transform
Skewness correction, i.e. one of
"none", "1/2", or "2/3".
y
Observed number
Value
Vector of length four with lower and upper bounds of an
(1−α)⋅100% confidence interval (first two
arguments) and corresponding quantile of observation y
together with the median of the predictive distribution.