Maximal First Differences for Generalized Linear Models

For objects of class glm, it calculates the change in predicted responses, for discrete changes in a covariate holding all other variables at their observed values.

glmChange2(obj, varname, data, change = c("unit", "sd"), R = 1500)

Arguments

obj

A model object of class glm.
varname

Character string giving the variable name for which average effects are to be calculated.
data

Data frame used to fit object.
change

A string indicating the difference in predictor values to calculate the discrete change. sd gives plus and minus one-half standard deviation change around the median and unit gives a plus and minus one-half unit change around the median.
R

Number of simulations to perform.

Value

res

A vector of values giving the average and 95 percent confidence bounds

ames

The average change in predicted probability (across all N observations) for each of the R simulations.

avesamp

The average change in predicted probability for each of the N observation (across all of the R simulations).

Details

The function calculates the average change in predicted probabiliy for a discrete change in a single covariate with all other variables at their observed values, for objects of class glm. This function works with polynomials specified with the poly function.

Examples


data(france)
left.mod <- glm(voteleft ~ male + age + retnat +
        poly(lrself, 2), data=france, family=binomial)
glmChange2(left.mod, "age", data=france, "sd")
#>           mean      lower       upper
#> age -0.0416124 -0.0665044 -0.01553706