Convert (refit) a Bayesian model to frequentist — convert_bayesian_as

Refit Bayesian model as frequentist. Can be useful for comparisons.

Usage

convert_bayesian_as_frequentist(
  model,
  data = NULL,
  REML = TRUE,
  response = NULL
)

bayesian_as_frequentist(model, data = NULL, REML = TRUE, response = NULL)

Arguments

model: A Bayesian model.
data: Data used by the model. If NULL, will try to extract it from the model. For multivariate brms models, rows with missing values are dropped as in brms, unless the model has addition terms such as y | mi().
REML: For mixed effects, should models be estimated using restricted maximum likelihood (REML) (TRUE, default) or maximum likelihood (FALSE)? This has no effect on refits with lme4::glmer(), or with glmmTMB::glmmTMB() when lme4::glmer() fails. Those always use maximum likelihood.
response: For multivariate brms models, the name of the one response to convert, as returned by names(insight::find_formula(model)). Other models ignore response.

Details

Linear models from BayesFactor with random factors (see whichRandom in BayesFactor::lmBF()) are refit with lme4::lmer(). A random term with only factors, such as cyl:gear, becomes (1 | cyl:gear). A random term with a continuous variable, such as Days:Subject, becomes (0 + Days | Subject).

Examples

# \donttest{
# Rstanarm ----------------------
# Simple regressions
model <- rstanarm::stan_glm(Sepal.Length ~ Species,
  data = iris, chains = 2, refresh = 0
)
bayesian_as_frequentist(model)
#> 
#> Call:
#> stats::lm(formula = formula$conditional, data = data)
#> 
#> Coefficients:
#>       (Intercept)  Speciesversicolor   Speciesvirginica  
#>             5.006              0.930              1.582  
#> 

model <- rstanarm::stan_glm(vs ~ mpg,
  family = "binomial",
  data = mtcars, chains = 2, refresh = 0
)
bayesian_as_frequentist(model)
#> 
#> Call:  stats::glm(formula = formula$conditional, family = family, data = data)
#> 
#> Coefficients:
#> (Intercept)          mpg  
#>     -8.8331       0.4304  
#> 
#> Degrees of Freedom: 31 Total (i.e. Null);  30 Residual
#> Null Deviance:	    43.86 
#> Residual Deviance: 25.53 	AIC: 29.53

# Mixed models
model <- rstanarm::stan_glmer(
  Sepal.Length ~ Petal.Length + (1 | Species),
  data = iris, chains = 2, refresh = 0
)
bayesian_as_frequentist(model)
#> Linear mixed model fit by REML ['lmerMod']
#> Formula: Sepal.Length ~ Petal.Length + (1 | Species)
#>    Data: data
#> REML criterion at convergence: 119.793
#> Random effects:
#>  Groups   Name        Std.Dev.
#>  Species  (Intercept) 1.0778  
#>  Residual             0.3381  
#> Number of obs: 150, groups:  Species, 3
#> Fixed Effects:
#>  (Intercept)  Petal.Length  
#>       2.5045        0.8885  

model <- rstanarm::stan_glmer(vs ~ mpg + (1 | cyl),
  family = "binomial",
  data = mtcars, chains = 2, refresh = 0
)
bayesian_as_frequentist(model)
#> Generalized linear mixed model fit by maximum likelihood (Laplace
#>   Approximation) [glmerMod]
#>  Family: binomial  ( logit )
#> Formula: vs ~ mpg + (1 | cyl)
#>    Data: data
#>       AIC       BIC    logLik -2*log(L)  df.resid 
#>   31.1738   35.5710  -12.5869   25.1738        29 
#> Random effects:
#>  Groups Name        Std.Dev.
#>  cyl    (Intercept) 1.925   
#> Number of obs: 32, groups:  cyl, 3
#> Fixed Effects:
#> (Intercept)          mpg  
#>     -3.9230       0.1723  
# }