Administrator
|
Your observation is very correct, Brice.
Since we always have a finite number of data, all those plots are essentially step functions.
However, it is also true that the step-wise patterns in C and D are relatively stronger than those of others.
So, we only need to find out the factor that makes the patterns stronger in C and D.
In statistics, a lot of times the answers are in raw data because those contain the whole information.
So, my suggestion is take a look the patterns of the raw data of C and D.
(This question might be a bit vague in that we have not learned way to draw the QQ-plot (by hands).
Let me briefly show this in the next lecture and it will clarify the vagueness.)
|