Pool Predictions or Estimated Marginal Means

Source: R/pool_predictions.R

This function "pools" (i.e. combines) multiple ggeffects objects, in a similar fashion as mice::pool().

pool_predictions(x, ...)

Arguments

x

A list of ggeffects objects, as returned by ggpredict, ggemmeans or ggeffect.
...

Currently not used.

Value

A data frame with pooled predictions.

Details

Averaging of parameters follows Rubin's rules (Rubin, 1987, p. 76).

References

Rubin, D.B. (1987). Multiple Imputation for Nonresponse in Surveys. New York: John Wiley and Sons.

Examples

# example for multiple imputed datasets
if (require("mice")) {
  data("nhanes2")
  imp <- mice(nhanes2, printFlag = FALSE)
  predictions <- lapply(1:5, function(i) {
    m <- lm(bmi ~ age + hyp + chl, data = complete(imp, action = i))
    ggpredict(m, "age")
  })
  pool_predictions(predictions)
}

#> Loading required package: mice
#> 
#> Attaching package: ‘mice’
#> The following object is masked from ‘package:stats’:
#> 
#>     filter
#> The following objects are masked from ‘package:base’:
#> 
#>     cbind, rbind
#> # Predicted values of bmi
#> # x = age
#> 
#> x     | Predicted |         95% CI
#> ----------------------------------
#> 20-39 |     29.11 | [27.15, 31.07]
#> 40-59 |     23.68 | [20.96, 26.41]
#> 60-99 |     22.62 | [18.98, 26.26]
#> 
#> Adjusted for:
#> * hyp =     no
#> * chl = 191.65