Often used to model positively skewed data when random variables are greater than 0. For example, the gamma distribution can describe the time for an electrical component to fail. Most electrical components of a given type will fail around the same time, but a few will take a long time to fail. The gamma distribution is commonly used in reliability survival studies.
The gamma distribution is defined by 3 parameters: shape, scale, threshold (for 3 parameter model).
|
|
In this graph, the gamma distribution is defined by different shape and scale values when the threshold is set at 0. As you can see from this graph, most values in a gamma distribution occur near each other, but some values trail into the upper tail. |
When the shape parameter is a positive integer, the gamma distribution is sometimes called an Erlang distribution. The Erlang distribution is used frequently in queuing theory applications.
