These functions calculate scores (CRPS, logarithmic score) and its gradient and Hessian with respect to the parameters of a location-scale transformed logistic distribution. Furthermore, the censoring transformation and the truncation transformation may be introduced on top of the location-scale transformed logistic distribution.
## score functions
crps_logis(y, location = 0, scale = 1)
crps_clogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
crps_tlogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
crps_gtclogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf, lmass = 0, umass = 0)
logs_logis(y, location = 0, scale = 1)
logs_tlogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
dss_logis(y, location = 0, scale = 1)## gradient (location, scale) functions
gradcrps_logis(y, location = 0, scale = 1)
gradcrps_clogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
gradcrps_tlogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
## Hessian (location, scale) functions
hesscrps_logis(y, location = 0, scale = 1)
hesscrps_clogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
hesscrps_tlogis(y, location = 0, scale = 1, lower = -Inf, upper = Inf)
For the score functions: a vector of score values.
For the gradient and Hessian functions: a matrix with column names corresponding to the respective partial derivatives.
vector of observations.
vector of location parameters.
vector of scale paramters.
lower and upper truncation/censoring bounds.
vectors of point masses in lower
and upper
respectively.