Behavioral Research Data Analysis with R by Yuelin Li & Jonathan Baron

Author:Yuelin Li & Jonathan Baron
Language: eng
Format: epub
Publisher: Springer New York, New York, NY

6.2.2 Checking Model Distributional Assumptions

A plot of model residuals against fitted values shows a problem in the model. Figure 6.2 shows that diamonds 116, 131, and 279 are outliers. They are identified automatically in a residual plot. The observed prices of these diamond stones greatly exceed the model’s predicted prices. The model underestimates the observed price of these larger stones with D and E color and VVS1 clarity (see the fitted column below). Diamonds of these characteristics are sold at a premium price, much higher than the fitted values of the model. The values of these residuals clearly depart from what are expected if they are normally-distributed.

Fig. 6.2Plots of model residuals. On the left is a residuals vs. fitted value plot to identify outliers of the model’s prediction, plotted with the plot.lm(lm1, which = 1) command. On the right is a plot of observed quantiles of the residuals against the theoretical quantiles of a standard normal distribution, plotted with plot.lm(lm1, which = 2). A dotted line is added showing where the observed quantiles should be if they follow a standard normal distribution. The two plots identifies outliers of model assumptions

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.