Estimation with and without outlier

The data provided in this exercise (hypdata_outlier) has an extreme outlier. A plot is shown of the dataset, and a linear regression model of response versus explanatory. You will remove the outlying point to see how one observation can affect the estimate of the line.

This exercise is part of the course

Inference for Linear Regression in R

View Course

Exercise instructions

Filter hypdata_outlier to remove the outlier.
Update the plot, p, to add another smooth layer (use geom_smooth).
- Like the other ribbon, the update should use the linear regression method, and not draw the ribbon.
- Unlike the other ribbon, the update should use the data = hypdata_no_outlier and be colored red.
- For now, just use the smooth curve, and not the confidence bounds (se = FALSE).

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# This plot is shown
p <- ggplot(hypdata_outlier, aes(x = explanatory, y = response)) + 
  geom_point() +
  geom_smooth(method = "lm", se = FALSE) 

# Filter to remove the outlier
hypdata_no_outlier <- ___

p +
  # Add another smooth lin .reg. layer, no ribbon, 
  # hypdata_no_outlier data, colored red
  ___

Edit and Run Code