Effect plots
One way to better understand the marginal effect of a unit change in price ratio is to use the logistic response model with some typical values and graph the predictions.
You can do this by using the function cplot() from the margins package. The function cplot() takes on the logistic.model object, the name of the predictor variable and the corresponding typical values as arguments. You assume the typical values for price.ratio to be in the range from -2 to 2. As you only want to investigate the predictions for a few values you define a sequence of numbers from -2 to 2 with increments 0.50 by using the function seq().
Deze oefening maakt deel uit van de cursus
Building Response Models in R
Oefeninstructies
- Define a sequence of numbers ranging from
-2to2and having increments0.50by using the functionseq(). Assign the result to an object namedx. - Plot the predicted purchase probabilities given
"price.ratio"andxby using the functioncplot().
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
# Define the sequence of x values
x <- seq(from = ___, to = ___, by = ___)
# Plot the price.ratio effect
___(logistic.model, ___, xvals = ___)