Description Usage Arguments Details Value References See Also Examples
This is a main function of this package. It performs the model averaging.
1 | table_model_averaging(models_list, models_list_boot, save_xlsx = FALSE)
|
models_list |
The list of fitted models |
models_list_boot |
The list of fitted models after bootstrap |
save_xlsx |
A boolean indicating whether saving the table as a *.xlsx document or not |
To compute averaged parameter estimates, we used the so-called "shrinkage estimator". That is, the parameter estimates from all models are being considered and parameters that were not estimated are considered to be equal to zero.
To compute the confidence intervals, we used the approximation [pnorm(0.025)*SE
; pnorm(0.975)*SE
].
To compute the standard error, we used equation 5 in Lukacs et al. 2010 (see References).
We indicated estimates higher than 1e-5 by a star stored in the last column of the table.
A data.frame storing the output of the model average.
Lukacs, Burnham & Anderson (2010). Model selection bias and Freedman’s paradox. Ann. Inst. Stat. Math. 62:117–125
boot_all_models
, table_boot
, list_tables_boot
, table_SE
1 2 | table_model_averaging(models_list = Models,
models_list_boot = Models_boot)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.