Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

common-distributions-one-pager

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

List of common distributions

Geometric(x|✓) = ✓(1 ✓)x 1(x 2 {0, 1, 2, . . .}) for 0 < ✓ < 1

Bernoulli(x|✓) = ✓x (1 ✓)1 x
1(x 2 {0, 1}) for 0 < ✓ < 1
✓ ◆
n x
Binomial(x|n, ✓) = ✓ (1 ✓)n x
1(x 2 {0, 1, . . . , n}) for 0 < ✓ < 1
x

e ✓ ✓x
Poisson(x|✓) = 1(x 2 {0, 1, 2, . . .}) for ✓ > 0
x!

✓x
Exp(x|✓) = ✓e 1(x > 0) for ✓ > 0

1
Uniform(x|a, b) = 1(a < x < b) for a < b
b a

ba a 1 bx
Gamma(x|a, b = rate) = x e 1(x > 0) for a, b > 0,
(a)

1
Gamma(x|a, b = scale) = xa 1
e x/b
1(x > 0) for a, b > 0,
ba (a)

↵c↵
Pareto(x|↵, c) = 1(x > c) for ↵, c > 0
x↵+1

1
Beta(x|a, b) = xa 1
(1 x)b 1
1(0 < x < 1) for a, b > 0
B(a, b)

2 1
N (x|µ, )= p exp 2
1
2 (x µ)2 for µ 2 R, 2 >0
2⇡ 2

r
1
N (x|µ, )= exp 1
2 (x µ)2 for µ 2 R, >0
2⇡

1 1 1
N (x|µ, C) = exp 2 (x µ)T C (x µ) for µ 2 Rd , C 2 Rd⇥d symmetric
(2⇡)d/2 |C|1/2
positive definite.

Other distributions
Suppose ⌃ ⇠ inverseWishart(vo , So 1 ), then
(vo +p+1)/2 1
p(⌃) / |⌃| exp { tr(So ⌃ )/2}.

You might also like