Beta distribution – beta
Second shape parameter, defaults to 1.
beta can be interpreted as the number of times (plus one) a one has been observed, so setting
beta to one means no observations,
beta=4 means 3 observations.
Likewise, the parameter
alpha can be interpreted as the number of times (plus one) a zero has been observed.
beta shape parameters are 1, the Beta distribution becomes a uniform distribution.
beta less than 1 lead to distributions that are highly skewed towards 0 and 1.
beta more than 1 lead to distributions that have mean
alpha/(alpha+beta) and zero density at 0 and 1.
See also Beta distribution.
Jordan Douglas’ distribution viewer: