translations for BCANZ report

[sebdalgarno]

currently only English

This report was generated through the [ssdtools Shiny app](https://bcgov-env.shinyapps.io/ssdtools)
which fits species sensitivity distributions to concentration data. The app is built from the R package [ssdtools](https://cran.r-project.org/web/packages/ssdtools/), and shares the same functionality.  

**Toxicant:** `r params$toxicant`   
**Report created:** `r format(Sys.Date(), "%d/%m/%Y")`  
**Input distributions:** `r params$dists`  
**ssdtools version:** `r utils::packageVersion("ssdtools")`  
**R version:** `r gsub("R version ", "", R.version$version.string)`   

\newpage

# Fit of all distributions
```{r, message=FALSE, warning=FALSE}
params$fit_plot

Goodness of fit table

kable(params$gof_table) 

Table 1. The goodness of fit statistics. 'dist' is the distribution name; 'ad' is the Anderson-Darling statistic; 'ks' is the Kolmogorov-Smirnov statistic; 'cvm' is the Cramer-von Mises statistic; 'aic' is the Akaike's Information Criterion; 'aicc' is the Akaike's Information Criterion corrected for sample size; 'bic' is the Bayesian Information Criterion; 'delta' is the Information Criterion differences; 'weight' is the Information Criterion weights. 'delta' and 'weight' are based on 'aic' for censored data and 'aicc' for non-censored data.

\newpage

Model averaged fit

params$model_average_plot 

Estimated hazardous/protective concentration

avehc <- ssd_hc(params$fit_dist, proportion = c(0.01, 0.05, 0.1, 0.2), ci = TRUE, nboot = params$nboot, min_pboot = 0.8, multi_est = TRUE)
avehc |>
  mutate(HCx = proportion*100,
         PCx = (1 - proportion) * 100) |> 
  dplyr::select(HCx, PCx, est, se, lcl, ucl, nboot, pboot) |>
  kable(digits = 3)

Table 2. The estimated hazardous/protective concentrations. 'HCx' is the % species affected; 'PCx' is the % species protected; 'est' is the model-averaged estimate of the concentration; 'se' is the bootstrap based standard error of the estimate; 'lcl' and 'ucl' are the bootstrapped based lower and upper 95% confidence limits; 'nboot' is the number of bootstrap samples; 'pboot' is the proportion of bootstrap samples that converged. The model-averaged estimate(s) is calculated by treating the distribution as a single weighted mixture distribution. Distributions with an absolute AIC difference greater than a default delta of 9.21 (corresponding to a weight of no more than 0.01) are excluded prior to bootstrapping.

\newpage

Input data

[sebdalgarno]

coming - will take 1-2 weeks

[sebdalgarno]

done