BiVariAn (version 1.0.3)

step_bw_p: Automatized stepwise backward for regression models (with forced terms)

Description

Automatized stepwise backward for regression models (with forced terms)

Usage

step_bw_p(
  reg_model,
  s_lower = "~1",
  s_upper = "all",
  trace = TRUE,
  steps = NULL,
  p_threshold = 0.05,
  data = NULL,
  forced = NULL,
  ...
)

Value

An object of class "step_bw" containing the final model and a data.frame of steps.

Arguments

reg_model

Regression model. Must be a glm or lm object

s_lower

Lower step. Names of the variables to be included at the lower step. Default is "~1" (intercept only)

s_upper

Upper step. Names of the variables to be included at the upper step. Default is "all" (all predictors)

trace

Logical. Whether to print each step to the console. Default is TRUE.

steps

Maximum number of elimination steps. If NULL, set to number of predictors.

p_threshold

P-value threshold for elimination. Default is 0.05.

data

Data frame for stepwise. If NULL, extracted from reg_model.

forced

Character vector of predictor names to always keep. Default NULL.

...

Additional arguments passed to car::Anova().

References

Efroymson MA. Multiple regression analysis. In: Ralston A, Wilf HS, editors. Mathematical methods for digital computers. New York: Wiley; 1960.

Examples

data(mtcars)
# Fit a linear model
regression_model <- lm(cyl ~ ., data = mtcars)
# Perform backward stepwise, forcing "wt" and "hp" to remain
stepwise <- step_bw_p(
  regression_model,
  forced = c("wt", "hp"),
  trace = FALSE
)
# Extract final model and view summary
final_model <- stepwise$final_model
summary(final_model)