## Beta distribution – alpha

First shape parameter, defaults to 1.
If both `alpha`

and `beta`

shape parameters are 1, the Beta distribution becomes a uniform distribution.

The parameter `alpha`

can be interpreted as the number of times (plus one) a zero has been observed, so setting `alpha`

to one means no observations, `alpha=4`

means 3 observations.
Likewise, the parameter `beta`

can be interpreted as the number of times (plus one) a one has been observed.

Values of `alpha`

and `beta`

less than 1 lead to distributions that are highly skewed towards 0 and 1.

Values of `alpha`

and `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: