R help - May 2009 - Stepwise logistic Regression with significance testing - stepAIC

Hello R-Users,
 
I have one binary dependent variable and a set of independent variables
(glm(formula,…,family=”binomial”) ) and I am using the function stepAIC (“MASS”)
for choosing an optimal model. However I am not sure if stepAIC considers
significance properties like Likelihood ratio test and Wald test (see example
below).  
 
> y <- rbinom(30,1,0.4)
> x1 <- rnorm(30)
> x2 <- rnorm(30)
> x3 <- rnorm(30)
> xdata <- data.frame(x1,x2,x3)
> 
> fit1 <- glm(y~ . ,family="binomial",data=xdata)
> stepAIC(fit1,trace=FALSE)
 
Call:  glm(formula = y ~ x3, family = "binomial", data = xdata) 
 
Coefficients:
(Intercept)           x3  
    -0.3556       0.8404  
 
Degrees of Freedom: 29 Total (i.e. Null);  28 Residual
Null Deviance:      40.38 
Residual Deviance: 37.86        AIC: 41.86 
> 
> fit <- glm( stepAIC(fit1,trace=FALSE)$formula 
,family="binomial")
> my.summ <- summary(fit)
> # Wald Test 
> print(my.summ$coeff[,4])
(Intercept)          x3 
  0.3609638   0.1395215 
> 
> my.anova <- anova(fit,test="Chisq")
> #LR Test
> print(my.anova$P[2])
[1] 0.1121783
>
 
 
Is there an alternative function or a possible way of checking if the added
variable and the new model are significant within the regression steps?
 
Thanks in advance for your help
 
Regards
 
Peter-Heinz Fox


      
	[[alternative HTML version deleted]]

Hello R-Users,
 
I have one binary dependent variable and a set of independent variables
(glm(formula,…,family=”binomial”) ) and I am using the function stepAIC (“MASS”)
for choosing an optimal model. However I am not sure if stepAIC considers
significance properties like Likelihood ratio test and Wald test (see example
below).  
 
> y <- rbinom(30,1,0.4)
> x1 <- rnorm(30)
> x2 <- rnorm(30)
> x3 <- rnorm(30)
> xdata <- data.frame(x1,x2,x3)
> 
> fit1 <- glm(y~ . ,family="binomial",data=xdata)
> stepAIC(fit1,trace=FALSE)
 
Call:  glm(formula = y ~ x3, family = "binomial", data = xdata) 
 
Coefficients:
(Intercept)           x3  
    -0.3556       0.8404  
 
Degrees of Freedom: 29 Total (i.e. Null);  28 Residual
Null Deviance:      40.38 
Residual Deviance: 37.86        AIC: 41.86 
> 
> fit <- glm( stepAIC(fit1,trace=FALSE)$formula 
,family="binomial")
> my.summ <- summary(fit)
> # Wald Test 
> print(my.summ$coeff[,4])
(Intercept)          x3 
  0.3609638   0.1395215 
> 
> my.anova <- anova(fit,test="Chisq")
> #LR Test
> print(my.anova$P[2])
[1] 0.1121783
>  
 
Is there an alternative function or a possible way of checking if the added
variable and the new model are significant within the regression steps?
 
Thanks in advance for your help
 
Regards
 
Peter-Heinz Fox



      
	[[alternative HTML version deleted]]

Hello R-Users,
 
I have one binary dependent variable and a set of independent variables
(glm(formula,…,family=”binomial”) ) and I am using the function stepAIC (“MASS”)
for choosing an optimal model. However I am not sure if stepAIC considers
significance properties like Likelihood ratio test and Wald test (see example
below).  
 
> y <- rbinom(30,1,0.4)
> x1 <- rnorm(30)
> x2 <- rnorm(30)
> x3 <- rnorm(30)
> xdata <- data.frame(x1,x2,x3)
> 
> fit1 <- glm(y~ . ,family="binomial",data=xdata)
> stepAIC(fit1,trace=FALSE)
 
Call:  glm(formula = y ~ x3, family = "binomial", data = xdata) 
 
Coefficients:
(Intercept)           x3  
    -0.3556       0.8404  
 
Degrees of Freedom: 29 Total (i.e. Null);  28 Residual
Null Deviance:      40.38 
Residual Deviance: 37.86        AIC: 41.86 
> 
> fit <- glm( stepAIC(fit1,trace=FALSE)$formula 
,family="binomial")
> my.summ <- summary(fit)
> # Wald Test 
> print(my.summ$coeff[,4])
(Intercept)          x3 
  0.3609638   0.1395215 
> 
> my.anova <- anova(fit,test="Chisq")
> #LR Test
> print(my.anova$P[2])
[1] 0.1121783
>  
 
Is there an alternative function or a possible way of checking if the added
variable and the new model are significant within the regression steps?
 
Thanks in advance for your help
 
Regards
 
Peter-Heinz Fox



      
	[[alternative HTML version deleted]]