Common

Expected
Univariate Cummulative Density Function Variance
Probability Mass Function (PMF) Value
Random (CDF) V(X)
E(X)
Variables
𝒇𝑿 (𝒙) = 𝒑𝒙 ∗ (𝟏 − 𝒑)(𝟏−𝒙)
0; when y < 0
𝑭𝑿 (𝒙) = 1 - p; 0 <= y < 1
𝑤ℎ𝑒𝑛 𝑥 = 1; 𝑓𝑋 (1) = 𝑝
1; y>=1
𝑤ℎ𝑒𝑛 𝑥 = 0; 𝑓𝑋 (0) = (1 − 𝑝)
Bernoulli V[X] = p * (1 - p)
Distributiuon
Variance is maximized at V[Y] =
X~Bernoulli(p) E[X] = p
25% when
p = 50%

𝒏
𝒇𝑿 (𝒙) = ( ) ∗ 𝒑𝒙 ∗ (𝟏 − 𝒑)(𝒏−𝒙)
𝒙
Binomial |𝒙|

Distribution 𝒏 𝒏! 𝒏
given that ( ) = 𝒇𝑿 (𝒙) = ∑ ( ) 𝒑𝒊 ∗ (𝟏 − 𝒑)𝒏−𝒊
𝒙 𝒙!(𝒏−𝒙)! 𝒊
B(n,p) 𝒊=𝟎
𝒏

measures the 𝑬[𝑿] = ∑ 𝒑

𝒊=𝟏
total number of
𝑽[𝑷] = 𝒏𝒑 ∗ (𝟏 − 𝒑)
successes from
an
independent 𝑬[𝑿] = 𝒏 ∗ 𝒑
Bernoulli
random
variable

|𝒙|
𝝀𝒊
𝝀𝒙 ∗ 𝒆−𝝀 𝑭𝑿 (𝒙) = 𝒆−𝝀 ∗ ∑
𝒇𝑿 (𝒙) = 𝒊!
𝒙! 𝒊=𝟎

Poisson
Y~Poisson(𝝀)

measures 𝑬[𝑿] = 𝝀 𝑽[𝑿] = 𝝀

count of events
over fixed time
spans

0 if x < a
𝒙−𝒂
𝟏 if a <= x <
𝒇𝑿 (𝒙) = 𝑭𝑿 (𝒙) 𝒃−𝒂
Uniform 𝒃−𝒂
= b
Distribution
If x >= b
assumes that (𝒃 − 𝒂)𝟐
𝒂+𝒃 𝑽[𝑿] =
any value 𝑬[𝑿] = 𝟏𝟐
𝟐
within the
range [a,b] is
equally likely
to occur
 ∞
Normal 𝟏 (−
(𝒙−𝝁)𝟐
)
𝒙− 𝝁 𝒙−𝝁 𝑬[𝑿] = 𝝁 𝐕[𝐗] = 𝝈𝟐
𝒇𝑿 (𝒙) = ∗ 𝒆 𝟐𝝈𝟐 𝑭𝑿 (𝒙) = 𝚽 ( )= ∫
Distribution √𝟐𝝅𝝈𝟐
𝝈 −∞ 𝝈
N(𝜇,𝜎) ∗ 𝝓(𝒛)

(𝒍𝒏(𝒙)−𝝁)𝟐 𝟏 𝟏 𝒍𝒏(𝒙) − 𝝁
𝟏 (−
𝟐𝝈𝟐
)
𝒇𝑿 (𝒙) = ∗ 𝒆 𝑭𝑿 (𝒙) = + 𝒆𝒓𝒇 [ ]
𝒙√𝟐𝝅𝝈𝟐 𝟐 𝟐 √𝟐𝝈𝟐

Log-normal
Distribution

𝟏 𝛎
( )−𝟏
𝒚
(− )
𝒇𝑿 (𝒙) = 𝛎 ∗ 𝒚 𝟐 ∗𝒆 𝟐
Type equation here.
( ) 𝛎
𝟐𝟐 ∗ 𝚪( )
𝟐

E[X] = 𝜈
𝑽[𝑿] = 𝛎𝟐
Chi-Squared
Distribution 𝜈: degrees of
𝜈: degrees of freedom
freedom

Type equation here. Type equation here.

E[X] = 0 𝛎
𝑽[𝑿] =
Student’s t 𝛎−𝟐
Distribution 𝜈: degrees of
freedom 𝜈: degrees of freedom

Type equation here.

Type equation here.

F Distribution
 𝟏 𝒙
(− ) 𝒙
𝒇𝑿 (𝒙) = ∗ 𝒆 𝜷 𝑭𝑿 (𝒙) = 𝟏 − 𝒆
(−𝜷)
𝜷

Exponential
E[X] = 𝛽 𝑽[𝑿] = 𝛃𝟐
Distribution

𝒙(𝜶−𝟏) (𝒙 − 𝟏)𝜷−𝟏
𝒇𝑿 (𝒙) =
𝑩(𝜶, 𝜷)

Beta 𝜶 𝜶∗𝜷
𝑬[𝑿] = 𝑽[𝑿] =
Distribution 𝜶+𝜷 (𝜶 + 𝜷)𝟐 ∗ (𝜶 + 𝜷𝜷 − 𝟏)

Type equation here.

Type equation here.

Mixed Normal
𝑽[𝑿] = 𝑬[𝑿𝟐 ] − 𝑬[𝑿]𝟐
Distribution