This branch implements both the weighted sample and uniroot methods for estimating hc values, as well as their confidence intervals.
The method used for model average is controlled through a new argument 'averaging_method' can take values of:
"arithmetic" - the original arithmetic weighted mean
"uniroot" - estimation of the weighted estimate and bootstrap CI's through the uniroot method
"weighted_sample" - estimation based on a weighted sample of bootstrap values
A future additions may include the option for taking a geometric weighted mean.
Test the various methods via:
data <- ssddata::ccme_boron
This branch implements both the weighted sample and uniroot methods for estimating hc values, as well as their confidence intervals.
The method used for model average is controlled through a new argument 'averaging_method' can take values of: "arithmetic" - the original arithmetic weighted mean "uniroot" - estimation of the weighted estimate and bootstrap CI's through the uniroot method "weighted_sample" - estimation based on a weighted sample of bootstrap values
A future additions may include the option for taking a geometric weighted mean.
Test the various methods via: data <- ssddata::ccme_boron
colnames(data) <- make.names(colnames(data))
dist <- ssd_fit_dists(data, left = 'Conc', dists = c('gamma', 'lgumbel', 'llogis', 'lnorm', 'weibull'), silent = TRUE, reweight = FALSE, min_pmix = 0, nrow = 6L, computable = TRUE, at_boundary_ok = FALSE, rescale = TRUE) ssd_hc(dist, ci = TRUE, nboot = 100, average = TRUE, averaging_method = "arithmetic")#root = FALSE, ssd_hc(dist, ci = TRUE, nboot = 10, average = TRUE, averaging_method = "uniroot", percent = c(5, 10)) ssd_hc(dist, ci = TRUE, nboot = 100, average = TRUE, averaging_method = "weighted_sample") ssd_hc(dist, ci = FALSE, nboot = 100, average = TRUE, averaging_method = "uniroot", percent = c(5, 10))