Build regression model from a set of candidate predictor variables by removing predictors based on p values, in a stepwise manner until there is no variable left to remove any more.

blr_step_p_backward(model, ...)

# S3 method for default
blr_step_p_backward(model, prem = 0.3, details = FALSE, ...)

# S3 method for blr_step_p_backward
plot(x, model = NA, print_plot = TRUE, ...)

Arguments

model

An object of class lm; the model should include all candidate predictor variables.

...

Other inputs.

prem

p value; variables with p more than prem will be removed from the model.

details

Logical; if TRUE, will print the regression result at each step.

x

An object of class blr_step_p_backward.

print_plot

logical; if TRUE, prints the plot else returns a plot object.

Value

blr_step_p_backward returns an object of class "blr_step_p_backward". An object of class "blr_step_p_backward" is a list containing the following components:

model

model with the least AIC; an object of class glm

steps

total number of steps

removed

variables removed from the model

aic

akaike information criteria

bic

bayesian information criteria

dev

deviance

indvar

predictors

References

Chatterjee, Samprit and Hadi, Ali. Regression Analysis by Example. 5th ed. N.p.: John Wiley & Sons, 2012. Print.

See also

Other variable selection procedures: blr_step_aic_backward(), blr_step_aic_both(), blr_step_aic_forward(), blr_step_p_forward()

Examples

if (FALSE) {
# stepwise backward regression
model <- glm(honcomp ~ female + read + science + math + prog + socst,
  data = hsb2, family = binomial(link = 'logit'))
blr_step_p_backward(model)

# stepwise backward regression plot
model <- glm(honcomp ~ female + read + science + math + prog + socst,
  data = hsb2, family = binomial(link = 'logit'))
k <- blr_step_p_backward(model)
plot(k)

# final model
k$model

}