Graphics Reference
In-Depth Information
Fig. 2.4
Example of non-normal distribution:
glass
data set for MLP: Histogram and Q-Q graphic
Fig. 2.5
Example of normal distribution:
pima
data set for MLP: Histogram and Q-Q graphic
Fig. 2.6
Example of a special case:
cleveland
data set for MLP: Histogram and Q-Q graphic
this fact. Finally, Fig.
2.6
shows a special case where the similarity between both
distributions, the sample of results and the normal distribution, is not confirmed by
all normality tests. In this case, one normality test could work better than another,
depending on types of data, number of ties or number of results collected. Due
to this fact, we have employed three well-known normality tests for studying the
normality condition. The choice of the most appropriate normality test depending on
the problem is out of the scope of this topic.
