ComenzarEmpieza gratis

Visualizing auto performance uncertainty

Confidence intervals are a very important visual indicator of model fit. Here you'll practice changing the appearance of confidence intervals and transforming the scale of partial effects plots.

Este ejercicio forma parte del curso

Nonlinear Modeling with Generalized Additive Models (GAMs) in R

Ver curso

Instrucciones del ejercicio

  • Plot the model (mod) that uses the mpg data, plotting only the partial effect of weight. Make the confidence interval shaded and "hotpink" in color.
  • Make another plot of the weight partial effect, this time shifting the scale by the value of the intercept using the shift argument, and including the uncertainty of the model intercept using the seWithMean argument.

Ejercicio interactivo práctico

Prueba este ejercicio completando el código de muestra.

library(mgcv)
# Fit the model
mod <- gam(hw.mpg ~ s(weight) + s(rpm) + s(price) + comp.ratio, 
           data = mpg, method = "REML")

# Plot the weight effect with colored shading
plot(mod, select = 1, ___)

# Make another plot adding the intercept value and uncertainty
plot(mod, select = 1, ___)
Editar y ejecutar código