Exam P Formula Sheet

FUNDAMENTALS CDFs AND DENSITIES COVARIANCES

• A′ = Ac = A complement • F (x) = P[X ≤ x] = CDF • Cov[X, Y ] = E[XY ] − E[X] · E[Y ]

• P [Ac ] = 1 − P [A] – Discrete X: • Cov[X, X] = Var[X]

P
• P [A ∪ B] = P [A]+P [B]−P [A ∩ B] F (x) = n≤x P[X = n] • Cov[aX + bY, cZ + dW ]
– Continuous X:
• Mutually exclusive if P [A ∩ B] = 0 Rx = ac Cov[X, Z] + ad Cov[X, W ]
F (x) = −∞
f (t) dt
+bc Cov[Y, Z] + bd Cov[Y, W ]
CONDITIONAL PROBABILITY d
• Density f (x) = F (x) • Var[aX + bY ] =
dx
P[A ∩ B] a2 Var[X]+2ab Cov[X, Y ]+b2 Var[Y ]
• P[A | B] = • Mixed distributions:
P[B] P
If F (x) has a jump at a, • Var [ Xi ]
• P[A ∩ B] = P[A] · P[B | A]
e.g., F (a) > lim F (x) = F (a− )
P
x↑a
= Var[Xi ] if independent
= P[B] · P[A | B]
Then P[X = a] = F (a) − F (a− )
• A and B are independent means: CONDITIONAL MOMENTS
– P[A ∩ B] = P[A] · P[B] (definition) MOMENTS • E[X | Y = y]
– P[A | B] = P[A] (implied) P
• Discrete: = x x · P[X = x | Y = y]
• A, B, and C are independent if ev- P
ery possible combination factors – E[X] = µ = x x · P[X = x] • E[X] = E[E[X | Y ]]
– E[X 2 ] = x x2 · P[X = x]
P
For example:
ORDER STATISTICS
P
– P[ABC ′ ] = P[A]P[B]P[C ′ ] – E[g(X)] = x g(x) · P[X = x]

– P[Ac B] = P[Ac ]P[B] • Continuous: • X1 , X2 , . . . , Xn iid random vari-

R ables,
• A1 , A2 , . . . , Ak are a partition if: – E[X] = µ = x · f (x) dx Y1 ≤ Y2 ≤ · · · ≤ Yn sorted version
– E[X 2 ] = x2 · f (x) dx
R
– Ai ∩ Aj = ∅, i ̸= j R • P[min{X1 , . . . , Xn } > x]
–
P
P[Ai ] = 1 – E[g(X)] = g(x) · f (x) dx
n
= P[Y1 > x] = (P[X > x])
• If A1 , A2 , . . . , Ak are a partition: • Mixed:
Find discrete / continuous pieces, • P[max{X1 , . . . , Xn } ≤ x]
– Law of total then sum
P probability: = P[Yn ≤ x] = (P[X ≤ x])
n
P[B] = P[B | Ai ] · P[Ai ]
• Survival approach: If X ≥ 0 then:
– Bayes’ Theorem: R∞ • fYi (y) = density of Yi =
P[Ai ∩ B] – E[X] = 0 S(x) dx
P[Ai | B] = i· ni [FX (y)]i−1 fX (y)[1−FX (y)]n−i


P[B] R∞
– E[g(X)] = 0 g ′ (x)S(x) dx
P[Ai ∩ B]
=P assuming g(0) = 0
P[B | Aj ] · P[Aj ]

JOINT DISTRIBUTIONS
VARIANCES P
• P[X = x] = y P[X = x, Y = y]
2 2 2
• Var[X] = σ = E[X ] − (E[X]) = Marginal distribution of X
• E[X 2 ] = Var[X] + (E[X])2 • P[X = x | Y = y]
2
• Var[cX] = c Var[X] P[X = x, Y = y]
=
p
• SD[X] = Var[X] P[Y = y]
= Conditional distribution of X
SD[X]
• Coefficient of Variation =
E[X] • FX,Y (x, y) = P[X ≤ x, Y ≤ y]
• FX (x) = P[X ≤ x] = FX,Y (x, ∞)

©2025 The Infinite Actuary [Link]fi[Link] Exam P Formula Sheet 1/3

 Exam P Formula Sheet
DISCRETE DISTRIBUTIONS

Variable Key Properties P[X = x] E[X] Var[X] mgf

1 n+1 n2 − 1
Uniform on {1, 2, . . . , n} ,1≤x≤n
n 2 12
0 or 1
Bernoulli p, x = 1 p p(1 − p) pet + q
1 − p, x = 0
 
Sum of Bernoullis n x
Binomial p (1 − p)n−x np np(1 − p) (pet + q)n
number of successes in n trials x

Number of trials 1 1−p pet

Geometric on {1, 2, . . .} p(1 − p)x−1
until first success p p2 1 − qet

Number of failures 1 1−p p

Geometric on {0, 1, . . .} p(1 − p)x −1
before first success p p2 1 − qet
 r
Sum of Geometric on 0, . . . x+r−1

r(1 − p) r(1 − p) p
Negative Binomial pr (1 − p)x
Number of failures until r-th success r−1 p p2 1 − qet
n t
Poisson Sum of Poissons is Poisson e−λ λn! λ λ eλ(e −1)

Bernoulli Variance Shortcut: If X is a random variable that can only take on 2 values, with P [X = a] = p and P [X = b] = 1 − p,
then Var(X) = (b − a)1 p(1 − p).

CONTINUOUS DISTRIBUTIONS
Densities are only given for the region on which they are positive, and are 0 otherwise.

Distribution range density cdf E[X] Var[X] MGF

1 1 et − 1
Uniform on (0, 1) 0<x<1 1 x for 0 < x < 1
2 12 t
1 x−a a+b (b − a)2 etb − eta
Uniform on (a, b) a<x<b
b−a b−a 2 12 t(b − a)

Exponential
with “rate” λ 1 −x/θ 1 1 1 λ
0<x<∞ λe−λx = e 1 − e−λx = 1 − e−x/θ θ= θ2 = =
1 θ λ λ2 1 − θt λ−t
and mean θ =
λ
α
xα−1 e−x/θ

1
Gamma 0<x<∞ see below αθ αθ2
θα (α − 1)! 1 − θt
1 2 2
Standard Normal −∞ < x < ∞ √ e−x /2 Φ(x) 0 1 et /2
2π
 
1 2 2 x−µ 2
σ 2 )/2
Normal(µ, σ 2 ) −∞ < x < ∞ √ e−(x−µ) /(2σ ) Φ µ σ2 eµt+(t
σ 2π σ

©2025 The Infinite Actuary [Link]fi[Link] Exam P Formula Sheet 2/3

 Exam P Formula Sheet
GAMMA SHORTCUT
One trick to remembering the Gamma distribution is to recall that if α is an integer, then a Gamma(α, θ) distribution is the sum
of α exponentials with mean θ.
• For α = 1, a Gamma is an exponential with mean θ and so has CDF P[X ≤ t] = 1 − e−t/θ
• For α = 2, a Gamma has CDF  
t
P[X ≤ t] = 1 − e−t/θ − e−t/θ
θ

• For α = 3, a Gamma has CDF

(t/θ)2
   
t
P[X ≤ t] = 1 − e−t/θ − e−t/θ − e−t/θ
θ 2

• Note that each time we increase α by 1, we subtract another term from the CDF that resembles a term from the probability
distribution of a Poisson. In general, if α is an integer, the CDF of a Gamma(α, θ) is given by:

P[X ≤ t] = P[X < t] = 1 − P[N ≤ α − 1] = P[N ≥ α]

where N is a Poisson random variable with mean λ = t/θ.

©2025 The Infinite Actuary [Link]fi[Link] Exam P Formula Sheet 3/3