Figure 3

Prediction error across variables. Note: Variables ordered by largest to smallest prediction error (left to right). Black and transparent dots indicate the distribution of individual data points for prediction error across variables. Black dots above the distribution of data points represent variables that show statistically significant differences, whereas red dots above the distribution of data points represent reference variables for comparing statistically significant differences between more than two variables. Error bars represent the 95% confidence interval of the mean. *** pā<ā.001.