table_boot: Build the table of the results of the non-parametric...

In courtiol/SileR: Survival Analysis of Elephants Using an Extension of the William Siler's Framework

Description

This function is called internally by other functions. It processes the outcomes of the fit of the same model on different bootstrap replicates.

Usage

table_boot(model, models_list_boot, print = TRUE)

Arguments

model	A fitted model
models_list_boot	The list of fitted models after bootstrap
print	A boolean indicating whether or not to print information about potential issues

Details

The table(s) produced contain the following columns:

1. beta the parameter estimate in the original fit

2. mean the mean of the parameter estimates across the fits on the bootstrapped data

3. bias the difference between the column mean and beta

4. SE the standard _error_ of the parameter estimates computed as the standard _deviation_ of the parameter estimates across the fits on the bootstrapped data

5. lwr the lower boundary of the 95% confidence interval approximated as pnorm(0.025)*SE

6. upr the upper boundary of the 95% confidence interval approximated as pnorm(0.975)*SE

Value

A data frame storing the results.

See Also

boot_all_models, list_tables_boot, table_model_averaging

Examples

index <- unlist(lapply(strsplit(names(Models_boot), "_"),
                function(i) i[1] == "mod1")) ## find which bootstrap belong to mod1
mod1_boot <- Models_boot[index]
table_boot(model = Models$mod1, models_list_boot = mod1_boot)

courtiol/SileR documentation built on May 16, 2020, 8:10 p.m.