Parameters from multiply imputed repeated analyses
Source:R/methods_mice.R
model_parameters.mira.RdFormat models of class mira, obtained from mice::width.mids(), or of
class mipo.
Usage
# S3 method for class 'mipo'
model_parameters(
model,
ci = 0.95,
exponentiate = FALSE,
p_adjust = NULL,
keep = NULL,
drop = NULL,
verbose = TRUE,
...
)
# S3 method for class 'mira'
model_parameters(
model,
ci = 0.95,
exponentiate = FALSE,
p_adjust = NULL,
keep = NULL,
drop = NULL,
verbose = TRUE,
...
)Arguments
- model
An object of class
miraormipo.- ci
Confidence Interval (CI) level. Default to
0.95(95%).- exponentiate
Logical, indicating whether or not to exponentiate the coefficients (and related confidence intervals). This is typical for logistic regression, or more generally speaking, for models with log or logit links. It is also recommended to use
exponentiate = TRUEfor models with log-transformed response values. For models with a log-transformed response variable, whenexponentiate = TRUE, a one-unit increase in the predictor is associated with multiplying the outcome by that predictor's coefficient. Note: Delta-method standard errors are also computed (by multiplying the standard errors by the transformed coefficients). This is to mimic behaviour of other software packages, such as Stata, but these standard errors poorly estimate uncertainty for the transformed coefficient. The transformed confidence interval more clearly captures this uncertainty. Forcompare_parameters(),exponentiate = "nongaussian"will only exponentiate coefficients from non-Gaussian families.- p_adjust
Character vector, if not
NULL, indicates the method to adjust p-values. Seestats::p.adjust()for details. Further possible adjustment methods are"tukey","scheffe","sidak"and"none"to explicitly disable adjustment foremmGridobjects (from emmeans).- keep
Character containing a regular expression pattern that describes the parameters that should be included (for
keep) or excluded (fordrop) in the returned data frame.keepmay also be a named list of regular expressions. All non-matching parameters will be removed from the output. Ifkeepis a character vector, every parameter name in the "Parameter" column that matches the regular expression inkeepwill be selected from the returned data frame (and vice versa, all parameter names matchingdropwill be excluded). Furthermore, ifkeephas more than one element, these will be merged with anORoperator into a regular expression pattern like this:"(one|two|three)". Ifkeepis a named list of regular expression patterns, the names of the list-element should equal the column name where selection should be applied. This is useful for model objects wheremodel_parameters()returns multiple columns with parameter components, like inmodel_parameters.lavaan(). Note that the regular expression pattern should match the parameter names as they are stored in the returned data frame, which can be different from how they are printed. Inspect the$Parametercolumn of the parameters table to get the exact parameter names.- drop
See
keep.- verbose
Toggle warnings and messages.
- ...
Arguments passed to or from other methods.
Details
model_parameters() for objects of class mira works
similar to summary(mice::pool()), i.e. it generates the pooled summary
of multiple imputed repeated regression analyses.
Examples
library(parameters)
data(nhanes2, package = "mice")
imp <- mice::mice(nhanes2)
#>
#> iter imp variable
#> 1 1 bmi hyp chl
#> 1 2 bmi hyp chl
#> 1 3 bmi hyp chl
#> 1 4 bmi hyp chl
#> 1 5 bmi hyp chl
#> 2 1 bmi hyp chl
#> 2 2 bmi hyp chl
#> 2 3 bmi hyp chl
#> 2 4 bmi hyp chl
#> 2 5 bmi hyp chl
#> 3 1 bmi hyp chl
#> 3 2 bmi hyp chl
#> 3 3 bmi hyp chl
#> 3 4 bmi hyp chl
#> 3 5 bmi hyp chl
#> 4 1 bmi hyp chl
#> 4 2 bmi hyp chl
#> 4 3 bmi hyp chl
#> 4 4 bmi hyp chl
#> 4 5 bmi hyp chl
#> 5 1 bmi hyp chl
#> 5 2 bmi hyp chl
#> 5 3 bmi hyp chl
#> 5 4 bmi hyp chl
#> 5 5 bmi hyp chl
fit <- with(data = imp, exp = lm(bmi ~ age + hyp + chl))
model_parameters(fit)
#> # Fixed Effects
#>
#> Parameter | Coefficient | SE | 95% CI | Statistic | df | p
#> -----------------------------------------------------------------------------
#> (Intercept) | 18.50 | 3.93 | [ 9.09, 27.91] | 4.70 | 6.61 | 0.003
#> age40-59 | -6.03 | 1.95 | [-10.36, -1.69] | -3.10 | 9.94 | 0.011
#> age60-99 | -7.62 | 2.47 | [-13.56, -1.68] | -3.08 | 6.51 | 0.020
#> hypyes | 2.42 | 1.99 | [ -2.02, 6.86] | 1.22 | 9.78 | 0.251
#> chl | 0.06 | 0.02 | [ 0.01, 0.11] | 2.62 | 6.81 | 0.035
#>
#> Uncertainty intervals (equal-tailed) and p-values (two-tailed)
#> computed using a Wald distribution approximation.
# \donttest{
# model_parameters() also works for models that have no "tidy"-method in mice
data(warpbreaks)
set.seed(1234)
warpbreaks$tension[sample(1:nrow(warpbreaks), size = 10)] <- NA
imp <- mice::mice(warpbreaks)
#>
#> iter imp variable
#> 1 1 tension
#> 1 2 tension
#> 1 3 tension
#> 1 4 tension
#> 1 5 tension
#> 2 1 tension
#> 2 2 tension
#> 2 3 tension
#> 2 4 tension
#> 2 5 tension
#> 3 1 tension
#> 3 2 tension
#> 3 3 tension
#> 3 4 tension
#> 3 5 tension
#> 4 1 tension
#> 4 2 tension
#> 4 3 tension
#> 4 4 tension
#> 4 5 tension
#> 5 1 tension
#> 5 2 tension
#> 5 3 tension
#> 5 4 tension
#> 5 5 tension
fit <- with(data = imp, expr = gee::gee(breaks ~ tension, id = wool))
#> Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27
#> running glm to get initial regression estimate
#> (Intercept) tensionM tensionH
#> 36.04762 -12.26984 -13.71429
#> Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27
#> running glm to get initial regression estimate
#> (Intercept) tensionM tensionH
#> 35.04545 -10.29545 -12.98295
#> Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27
#> running glm to get initial regression estimate
#> (Intercept) tensionM tensionH
#> 35.150000 -8.973529 -13.267647
#> Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27
#> running glm to get initial regression estimate
#> (Intercept) tensionM tensionH
#> 36.66667 -13.26667 -14.50000
#> Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27
#> running glm to get initial regression estimate
#> (Intercept) tensionM tensionH
#> 36.15000 -11.37222 -14.21250
# does not work:
# summary(mice::pool(fit))
model_parameters(fit)
#> New names:
#> • `` -> `...6`
#> New names:
#> • `` -> `...6`
#> New names:
#> • `` -> `...6`
#> New names:
#> • `` -> `...6`
#> New names:
#> • `` -> `...6`
#> # Fixed Effects
#>
#> Parameter | Coefficient | SE | 95% CI | Statistic
#> -----------------------------------------------------
#> (Intercept) | 35.81 | 2.71 | | 13.21
#> tensionM | -11.24 | 4.31 | | -2.61
#> tensionH | -13.74 | 3.98 | | -3.45
#>
#> Uncertainty intervals (equal-tailed) and p-values (two-tailed)
#> computed using a Wald distribution approximation.
# }
# and it works with pooled results
data("nhanes2", package = "mice")
imp <- mice::mice(nhanes2)
#>
#> iter imp variable
#> 1 1 bmi hyp chl
#> 1 2 bmi hyp chl
#> 1 3 bmi hyp chl
#> 1 4 bmi hyp chl
#> 1 5 bmi hyp chl
#> 2 1 bmi hyp chl
#> 2 2 bmi hyp chl
#> 2 3 bmi hyp chl
#> 2 4 bmi hyp chl
#> 2 5 bmi hyp chl
#> 3 1 bmi hyp chl
#> 3 2 bmi hyp chl
#> 3 3 bmi hyp chl
#> 3 4 bmi hyp chl
#> 3 5 bmi hyp chl
#> 4 1 bmi hyp chl
#> 4 2 bmi hyp chl
#> 4 3 bmi hyp chl
#> 4 4 bmi hyp chl
#> 4 5 bmi hyp chl
#> 5 1 bmi hyp chl
#> 5 2 bmi hyp chl
#> 5 3 bmi hyp chl
#> 5 4 bmi hyp chl
#> 5 5 bmi hyp chl
fit <- with(data = imp, exp = lm(bmi ~ age + hyp + chl))
pooled <- mice::pool(fit)
model_parameters(pooled)
#> # Fixed Effects
#>
#> Parameter | Coefficient | SE | 95% CI | Statistic | df | p
#> -------------------------------------------------------------------------------
#> (Intercept) | 19.05 | 3.39 | [ 11.82, 26.28] | 5.61 | 15.19 | < .001
#> age40-59 | -4.97 | 1.86 | [ -9.02, -0.92] | -2.67 | 12.12 | 0.020
#> age60-99 | -6.14 | 1.89 | [-10.15, -2.12] | -3.25 | 15.35 | 0.005
#> hypyes | 2.11 | 2.29 | [ -3.38, 7.59] | 0.92 | 6.60 | 0.390
#> chl | 0.05 | 0.02 | [ 0.01, 0.09] | 2.83 | 15.74 | 0.012
#>
#> Uncertainty intervals (equal-tailed) and p-values (two-tailed)
#> computed using a Wald distribution approximation.