The ‘shape’ of the errors
The spread of a set of values (errors) can follow different probability distributions.
Normal or Gaussian distribution
Sometimes the values are more likely to fall near the average than further away. You might see this type of distribution if you examined the heights of individuals in a large group of people. Most people are close to average height; few are extremely tall or short.
The ‘blob plot’ shows a set of values lying in a normal
distribution. General sketch of a normal distribution.
distribution. General sketch of a normal distribution.
Uniform or rectangular distribution
Sometimes values are equally likely to be evenly spread between the highest and the lowest values. An example of this would be seen if you examined how raindrops fall on a thin, straight telephone wire. They would be as likely to fall on any one part as on another.
The blob plot shows a set of values lying in a rectangular distribution and general sketch of a rectangular distribution.
7 of 9