Exam FAM

ACTEX Learning Formula & Review Sheet

(updated 07/31/2024)

FAM-S
A. REVIEW OF PROBABILITY
A1. BASIC PROBABILITY

Cumulative Distribution Function: F (x) = Pr (X ≤ x)

Survival Function: S (x) = 1−F (x) = Pr (X > x)

Probability Density Function: f (x) = d

dx F
d
(x) = − dx S (x)

Hazard Rate Function: log S (x)

Rx
λ (x) = f (x)
S(x)
d
= − dx → S (x) = e− −∞
λ(t)dt

Cumulative Hazard Function:

Rx
Λ (x) = −∞
λ (t) dt → S (x) = e−Λ(x)

dg −1 (y)
Transformation: fY (y) = fX g −1 (y)


Y = g (X) → dy

A2. EXPECTATION & VARIANCE

Expectation (Discrete X): x Pr (X = x) g (x) Pr (X = x)

P P
E [X] = → E [g (X)] =

Expectation (Continuous X):

R∞ R∞
E [X] = −∞
xf (x) dx → E [g (X)] = −∞
g (x) f (x) dx

Variance:
  2
V ar (X) = E X 2 − E [X]

Standard Deviation:
p
SD (X) = V ar (X)

Covariance: Cov (X, Y ) = E [XY ] − E [X] E [Y ]

Cov (X, Y )
Correlation Coefficient: Corr (X, Y ) =
SD (X) SD (Y )

Linear Combination: W = aX + bY → E [W ] = aE [X] + bE [Y ]

→ V ar (W ) = a2 V ar (X) + b2 V ar (Y )

+ 2ab Cov (X, Y )

Sum of iid X: S = X1 + · · · + Xn → E [S] = nE [X]

→ V ar (S) = nV ar (X)

Need Help? Email [email protected] Copyright © ArchiMedia Advantage Inc. All Rights Reserved
 ACTEX Learning Exam FAM Page 2

A3. CONDITIONAL PROBABILITY

Pr (X=x, Y =y) Pr (X=x) Pr (Y =y|X=x)

Conditional Probability: Pr (X = x|Y = y) = Pr (Y =y) = Pr (Y =y)

Conditional Density Function: f (x|y) = f (x,y)

f (y) = f (x)f (y|x)
f (y)

Conditional expectation (Discrete): x Pr (X = x|Y = y)

P
E [X|Y = y] =

Conditional expectation (Continuous):

R∞
E [X|Y = y] = −∞
xf (x|y)dx

Law of Total Probability: Pr (X ≤ x) = E [Pr (X ≤ x|Y )]

Law of Total Expectation: E [X] = E [E [X|Y ]]

Law of Total Variance: V ar (X) = E [V ar (X|Y )] + V ar (E [X|Y ])

B. SEVERITY, FREQUENCY
& AGGREGATE MODELS
B1. SEVERITY DISTRIBUTIONS

Distribution Probability density function Formulas worth memorizing

1 x−a a+b (b − a)2

Uniform f (x) = , a≤x≤b F (x) = E[X] = V ar(X) =
b−a b−a 2 12
1 −x
Exponential
x
f (x) = e θ, x > 0 F (x) = 1 − e− θ E[X] = θ V ar(X) = θ2
θ
τ  x τ − x
τ
Weibull
x τ


f (x) = e θ , x>0 F (x) = 1 − e− θ
x θ

F (x) = Pr (X ∗ ≥ α)
1 α


Gamma
x
f (x) = θ
xα−1 e− θ , x > 0 x E [X] = αθ V ar (X) = αθ2
Γ (α) X ∗ is Poisson with λ =
θ
If α is an integer.

Γ (a + b) a−1 b−1 a   a (a + 1)
f (x) = x (1 − x) , E [X] = E X2 =
Beta Γ (a) Γ (b) a+b (a + b) (a + b + 1)

0<x<1 If a and b are integers. If a and b are integers.

θ
αθα

θ
α E [X] =
Pareto f (x) = α+1 , x>0 F (x) = 1 − α−1
(x + θ) x+θ
If α > 1 is an integer.
 α
αθα θ αθ
Single P. Pareto f (x) = , x>θ F (x) = 1 − E [X] =
xα+1 x α−1
log x − µ
 
1 (log x−µ)2 σ2
Lognormal
2
√ e− 2σ2 , x > 0
 
f (x) = F (x) = N E [X] = eµ+ 2 E X 2 = e2µ+2σ
xσ 2π σ

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 3

B2. FREQUENCY DISTRIBUTIONS

Distribution Probability mass function Formulas worth memorizing

Poisson P (x) = e−λ λx

x! , x = 0, 1, 2, . . . , ∞ E [N ] = λ V ar (N ) = λ

Binomial m

m−x
P (x) = x q x (1 − q) , x = 0, 1, 2, . . . , m E [N ] = mq V ar (N ) = mq (1 − q)
  x
Geometric P (x) = 1
1+β
β
1+β , x = 0, 1, 2, . . . , ∞ E [N ] = β V ar (N ) = β (1 + β)

 r  x
Negative Binomial P (x) = x+r−1
x
1
1+β
β
1+β , x = 0, 1, 2, . . . , ∞ E [N ] = rβ V ar (N ) = rβ (1 + β)

Zero-truncated distributions.: pTx = px

1−p0

 
Zero-modified distributions.: pM M
x = 1 − p0
px
1−p0

B3. AGGREGATE MODELS

Individual risk model: S = X1 + X2 + · · · + Xn → E[S] = nE[X]

→ V ar(S) = nV ar(X)

Collective risk model: S = X1 + X2 + · · · + XN → E[S] = E[N ]E[X]

→ V ar(S) = E[N ]V ar(X) + V ar(N )E[X]2

Normal Distribution:
iid


Xi ∼ N µ, σ 2 → S L = X1 + X2 + · · · + Xn ∼ N nµ, nσ 2

Exponential Distribution:
iid
Xi ∼ Exp (θ) → S L = X1 + X2 + · · · + Xn ∼ Gamma (α = n, θ)
 
Normal approximation: k−µ


S ∼N
˙ µ = E[S], σ 2 = V ar(S) → P r(S ≤ k) ≈ N σ

B4. MEASURES OF RISKS

Coherent risk measures: 1. Translation Invariance: ρ (X + c) = ρ (X) + c

2. Positive Homogeneity: ρ (cX) = cρ (X)

3. Subadditivity: ρ (X + Y ) ≤ ρ (X) + ρ (Y )

4. Monotonicity: ρ (X) ≤ ρ (Y ) If Pr (X ≤ Y ) = 1

Value-at-Risk: VaRα (X) Where Pr (X ≤ VaRα (X)) = α.

Tail-Value-at-Risk: TVARα (X) = E [X|X > VaRα (X)] = VaRα (X) + E[X]−E[X∧VaRα (X)]
1−α

E X2
 
Equilibrium Distribution: fe (x) = E[X] ,
S(x)
X>0 → E [Xe ] = 2E[X]

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 4

C. COVERAGE MODIFICATIONS
C1. PAYMENT PER LOSS

Policy Payment per loss Expected payment per loss


0,

X<d
With ordinary deductible d
 
YL = E Y L = E[X] − E[X ∧ d]
X − d,

X≥d

0, X ≤ d∗

With franchise deductible d∗ E Y L = E [X | X > d∗ ]
 
YL =
X, X > d∗



X, X≤u

With maximum covered loss u
 
YL = E Y L = E[X ∧ u]
u,

X>u




 0, X≤d


With d and u
 
YL = X − d, d < X ≤ u E Y L = E[X ∧ u] − E[X ∧ d]




u − d, X > u




 0, X≤d


With d, u and coinsurance factor α
 
YL = α(X − d), d < X ≤ u E Y L = α(E[X ∧ u] − E[X ∧ d])




α(u − d), X > u


d


 0, X ≤ 1+r

  h i h i
With d, u, α and inflation rate r
 
u d
 
YL = α(1 + r) X − d , d u E Y L = α(1 + r) E X ∧ −E X ∧
 1+r 1+r < X ≤ 1+r 1+r 1+r

  
α(1 + r) u − d , X > u

1+r 1+r 1+r

h i
E YL
Loss elimination ratio: LER = 1 − E[X]

C2. PAYMENT PER PAYMENT

h i
E YL
Payment per Payment P L
 
Y = Y |X > d → E YP = Pr (X>d)

Policy Loss Payment per payment

X ∼ Unif (0, b) Y P ∼ Unif (0, b − d)

With ordinary deductible d X ∼ Exp (θ) Y P ∼ Exp (θ)

X ∼ Pareto (α, θ) Y P ∼ Pareto (α, θ + d)

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 5

C3. REINSURANCE

Proportional reinsurance Insurer pays Reinsurer pays

With quota share α Y = (1 − α)X Y = αX

 
X, X≤u 0, X≤u
 
With retention u and surplus share α Y = Y =
u + (1 − α)(X − u), X > u
 α(X − u), X > u


Excess of loss reinsurance Insurer pays Reinsurer pays

 
X, X ≤ d
 0,

X≤d
Covers losses above d Y = Y =
d, X > d
 X − d, X > d


 


 X, X≤d 

0, X≤d

 

Covers losses above d but below u Y = d, d≤X<u Y = X − d, d ≤ X < u

 


 

X + d − u, X > u u − d, X > u

D. MAXIMUM LIKELIHOOD ESTIMATION

D1. MLE WITH COMPLETE DATA

For distributions that belong to the exponential family:

1. Determine L(θ).
2. Apply natural logarithm, obtain l(θ) = log L(θ).
3. Take the first derivative with respect to the parameter, obtain l0 (θ).
4. Set l0 (θ) = 0, obtain θ̂, which is the MLE.

Distribution Likelihood Function Maximum likelihood estimate(s)

Exponential θ̂ = x̄

Normal 1 2
σ̂ 2 =
P
µ̂ = x̄ n (xi − µ̂)
L(θ) = f (x1 ) . . . f (xn )
Lognormal log xi
1 1 2
σ̂ 2 =
P P
µ̂ = n n (log xi − µ̂)

Uniform bb = max (x1 , . . . , xn )

Binomial q̂ = x̄
m
L(θ) = p (x1 ) . . . p (xn )
Poisson λ̂ = x̄

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 6

D2. MLE WITH INCOMPLETE DATA

Likelihood function Note

Grouped data Where c0 < c1 < · · · < cn are interval boundaries.

m1 mn
L = (F (c1 ) − F (c0 )) . . . (F (cn ) − F (cn−1 ))

f (x1 ) f (xn )
Left-truncated data L= ... Losses below d are not reported.
S(d) S(d)

Right-censored data L = f (x1 ) . . . f (xn ) S(u)m Losses are capped at u.

Left-truncated
 m
f (x1 ) f (xn ) S(u)
L= ... Losses below d are not reported. Losses are capped at u.
S(d) S(d) S(d)
& Right-censored data

E. CLASSICAL CREDIBILITY
E1. FULL CREDIBILITY

We want...to be within k of Range parameter, k Total number of exposures Total number of claims
1 + p
the mean p of the time. Note: z = N −1 needed, eF needed, nF
2

Var(N )
r
z
eF
Average number of claims k=  z 2 Var(N )  z 2 Var(N )
E[N ] eF = × nF = × × E[N ]
k E[N ]2 k E[N ]2
z eF Var(N )
p
Total number of claims k=
eF E[N ]
Var(X)
r
z
nF
 z 2 Var(X) 1  z 2 Var(X)
Average claim size k= eF = × × nF = ×
E[X] k E[X]2 E[N ] k E[X]2
Var(S)
r
z
eF
Average aggregate claims k=  z 2 Var(S)  z 2 Var(S)
E[S] eF = × nF = × × E[N ]
k E[S]2 k E[S]2
z eF Var(S)
p
Total aggregate claims k=
eF E[S]

E2. POISSON FREQUENCY

We want...to be within k Total number of exposures needed, Total number of claims needed,
Frequency
of the mean p of the time. eF nF

Average number of claims  z 2 1  z 2

eF = × nF =
k λ k
Total number of claims

N ∼ Poisson(λ)
 z 2 Var(X) 1  z 2 Var(X)
Average claim size eF = × × nF = ×
k E[X]2 λ k E[X]2

Average aggregate claims  z 2 

Var(X)

1  z 2 
Var(X)

eF = × 1+ × nF = × 1+
k E[X]2 λ k E[X]2
Total aggregate claims

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 7

E3. PARTIAL CREDIBILITY

Number of exposures available

r
Credibility factor: Z=
Number of exposures needed for full credibility

Number of claims available

r
or Z=
Number of claims needed for full credibility

Credibility premium: P = Z × Observation + (1 − Z) × Manual Rate

F. RATEMAKING & LOSS RESERVING

F1. RATEMAKING DATA

Written exposure
Earned exposure
Exposure
Unearned exposure

Written premium
Earned premium
Premium
Unearned premium

Reported claims
Claim
Unreported claims

Reported loss = Paid loss + Case reserve

IBNR reserve = Incurred but not reported reserve
Loss
IBNER reserve = Incurred but not enough reported reserve
Ultimate loss = Reported loss + IBNR reserve + IBNER reserve

ALAE = Allocated loss adjustment expenses

Expenses
ULAE = Unallocated loss adjustment expenses

Number of Claims
Basic formulas: Frequency =
Number of Exposures
Losses
Severity =
Number of Claims
Losses
Pure Premium/Loss Cost = = Frequency × Severity
Number of Exposures
Losses
Loss Ratio =
Premium
Expenses
Expense Ratio =
Premium

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 8

F2. AGGREGATION METHODS

Aggregation methods Premium/Exposure Loss

Calendar year Transaction date Transaction date

Calendar-accident year Transaction date Accident date

Policy year Effective date Effective date

F3. LOSS RESERVING METHODS

1. Ultimate losses = Earned Premium × Expected Loss Ratio

Expected loss ratio method
2. Loss Reserve = Ultimate Losses − Paid Losses

1. Prepare a run-off triangle for paid losses.

2. Calculate age-to-age factors using average factor method or mean factor method.

3. Calculate age-to-ultimate factor fULT , which is the product of age-to-age factors.

Chain-ladder method
4. Ultimate Losses = Paid Losses × fULT

5. Loss reserve = Ultimate Losses − Paid Losses

1. Prepare a run-off triangle for paid losses.

2. Calculate age-to-age factors using average factor method or mean factor method.
Bornhuetter-Ferguson method
3. Calculate age-to-ultimate factor fULT , which is the product of age-to-age factors.
 
4. Ultimate Losses = Paid Losses + Earned Premium × Expected Loss Ratio × 1 − 1
fULT

5. Loss reserve = Ultimate Losses − Paid Losses

F4. PRICING FORMULA

General formula: Premium = Losses + Loss adjustment expenses + Fixed Expenses + Variable Expenses + Profit

L + LAE + F
Premium: P = L + LAE + F + (V + Q)P → P =
1−V −Q
L + LAE + F
Permissible loss ratio: R=1−V −Q → P =
R
Current rate level
Adjustments to data: Premium at current rates = Earned premium ×
Historical average rate level

Ultimate losses = Reported losses × Development factor

Trended losses = Reported losses × Trend factor

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 9

F5. RATEMAKING METHODS

Trended and ultimate losses and LAE

Projected loss cost including LAE =
Number of earned exposures
Loss cost method Projected loss cost+Fixed expenses per exposure
Indicated rate =
Permissible loss ratio
Indicated rate
Indicated rate change = −1
Current rate
Trended and ultimate losses and LAE
Projected loss and LAE ratio =
Earned premiums at current rate level
Loss ratio method Projected loss and LAE ratio + Fixed expense ratio
Indicated rate change = −1
Permissible loss ratio
Earned premiums at current rate level
Indicated rate = × (1 + Indicated rate change )
Number of earned exposures

G. Option Pricing
G1. PUT AND CALL OPTIONS

Financial derivative Payoff Profit

C(T ) = max(0, S(T ) − K) Profit = C(T ) − C(0)erT

Call option

S(T ) S(T )
k
k

P (T ) = max(0, K − S(T )) Profit = P (T ) − P (0)erT

Put option

S(T ) S(T )
k
k

Put-call parity: C(0) − P (0) = S(0) − Ke−rT

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 10

G2. BINOMIAL OPTION PRICING MODEL

Stock price at time h: Su = S(0) × u or Sd = S(0) × d

Option payoff at time h: Vu or Vd

 
Vu − Vd uVu − dVu
Replicating portfolio: ∆= & B=e −rh
Su − Sd u−d

Option price: V (0) = ∆S(0) + B

erh − d
Risk neutral probability: p∗ =
u−d

Option price: V (0) = e−rh (p∗ Vu + (1 − p∗ ) Vd )

G3. BLACK-SCHOLES-MERTON MODEL

 2
 √
r− σ2
Stock price at time T : where Z ∼ N (0, 1)
T +σ T Z
S(T ) = S(0)e

Call price: C(0) = S(0)N (d1 ) − Ke−rT N (d2 )

Call delta: ∆C = N (d1 )

Put price: P (0) = Ke−rT N (−d2 ) − SN (−d1 )

Put delta: ∆p = −N (−d1 )

 
log S(0)
K + r+ σ2
2 T √
Where: d1 = √ & d2 = d1 − σ T
σ T

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 11

FAM-L
A. REVIEW OF PROBABILITY
A1. BASIC PROBABILITY

Cumulative Distribution Function: F (x) = Pr (X ≤ x)

Survival Function: S (x) = 1−F (x) = Pr (X > x)

Probability Density Function: f (x) = d

dx F
d
(x) = − dx S (x)

Hazard Rate Function: log S (x)

Rx
λ (x) = f (x)
S(x)
d
= − dx → S (x) = e− −∞
λ(t)dt

Cumulative Hazard Function:

Rx
Λ (x) = −∞
λ (t) dt → S (x) = e−Λ(x)

A2. EXPECTATION & VARIANCE

Expectation (Discrete X): x Pr (X = x) g (x) Pr (X = x)

P P
E [X] = → E [g (X)] =

Expectation (Continuous X):

R∞ R∞
E [X] = −∞
xf (x) dx → E [g (X)] = −∞
g (x) f (x) dx

Variance:
  2
V ar (X) = E X 2 − E [X]

Standard Deviation:
p
SD (X) = V ar (X)

Covariance: Cov (X, Y ) = E [XY ] − E [X] E [Y ]

Cov (X, Y )
Correlation Coefficient: Corr (X, Y ) =
SD (X) SD (Y )

Linear Combination: W = aX + bY

E [W ] = aE [X] + bE [Y ]

V ar (W ) = a2 V ar (X) + b2 V ar (Y ) + 2ab Cov (X, Y )

Sum of iid X: S = X1 + · · · + Xn

E [S] = E [X1 + · · · + Xn ] = nE [X]

V ar (S) = V ar (X1 + · · · + Xn ) = nV ar (X)

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 12

A3. CONDITIONAL PROBABILITY

Conditional expectation x Pr(X = x | Y = y)

P
E[X | Y = y] =
(Discrete):

Conditional expectation
R∞
E[X | Y = y] = −∞
xf (x | y)dx
(Continuous):

Law of total expectation: E[X] = E[E[X|Y ]]

Law of Total Variance: V ar (X) = E [V ar (X|Y )] + V ar (E [X|Y ])

A4. COMMON DISTRIBUTIONS

Uniform distribution: fX (x) = 1

b−a for a ≤ x ≤ b → E [X] = a+b
2 V ar (X) = (b−a)2
12

Exponential distribution: fX (x) = θ1 e− θ for x > 0

x
→ E [X] = θ V ar (X) = θ2
 
Normal approximation: Pr (S ≤ k) ≈ N
.
k−µ
S ∼ µ = E [S] , σ 2 = V ar (S) → σ

B. REVIEW OF FINANCIAL MATHEMATICS

B1. INTEREST RATES

Annual effective interest rate: i

Rate of discount: d= i
1+i = iv = 1 − v

Discounting rate: v= 1
1+i =1−d

Continuously compounded interest rate: δ = log (1 + i)

 
Nominal interest rate:
1
i(m) = m (1 + i) m − 1
 
Nominal rate of discount:
1
d(m) = m 1 − (1 − d) m

B2. PRESENT VALUES

PV of n-year certain annuity-due: ä n = 1−v n

d (1 at the beginning of each year)

PV of n-year certain annuity-immediate: an = 1−v n

i (1 at the end of each year)

PV of n-year continuous certain annuity: ā n = 1−v n

δ (1 per year continuously)

PV of n-year 1/m-thly certain annuity-due: 1−v n

(1/m at the beginning of each of the nm periods)
(m)
ä n = d(m)

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 13

C. SURVIVAL MODELS
C1. SURVIVAL FUNCTION

Complete future lifetime: Tx → The pdf of Tx is fx (t) = t px µx+t .

Curtate future lifetime: Kx = bTx c → The pmf of Kx is Pr (Kx = k) = k px qx+k .

Survival function: Sx (t) = Pr (Tx > t)

Sx (u + t)
Formulas: Sx (u + t) = Sx (u) Sx+u (t) → Sx+u (t) =
Sx (u)
S0 (x + t)
S0 (x + t) = S0 (x) Sx (t) → Sx (t) =
S0 (x)

Three conditions: (1) Sx (0) = 1

(2) lim Sx (t) = 0

t→∞

(3) Sx (t) must be a non-increasing function of t.

C2. ACTUARIAL NOTATION

Survival probability: t px = Pr (Tx > t)

Mortality probability: t qx = 1 − t px = Pr (Tx ≤ t)

Formulas: t+u px = t px u px+t

t+u qx = t qx + t px u qx+t

t|u q x = t px u qx+t = t px − t+u px = t+u qx − t qx

C3. LIFE TABLES

Number of lives: lx

Number of deaths: t dx = lx − lx+t

lx+t
Formulas: t px =
lx
d
t x lx − lx+t
t qx = =
lx lx
u dx+t lx+t − lx+t+u
t|u qx = =
lx lx

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 14

C4. FORCE OF MORTALITY

Definition: µx (t) = µx+t = fx (t)

Sx (t)
d
= − dt log Sx (t)

Condition: lim
Rt
µs ds = ∞
t→∞ 0

Formulas:
Rt R x+t
t px = e− 0
µx+s ds
= e− x
µs ds

Rt
t qx = p
0 s x
µx+s ds
R t+u
t|u qx = t s px µx+s ds

C5. EXPECTED FUTURE LIFETIME

Expectations:
o R∞ R∞
E [Tx ] = ex = 0
t t px µx+t dt = 0 t px dt

∞
P ∞
P
E [Kx ] = ex = k k px qx+k = k px
k=1 k=1
o Rn Rn
E [min (Tx , n)] = ex:n = 0
t t px µx+t dt + nn px = p dt
0 t x
n−1
P n
P
E [min (Kx , n)] = ex:n = k k px qx+k + nn px = k px
k=1 k=1

Second moments:
  R∞ R∞
E Tx2 = 0 t2 t px µx+t dt = 0 2tt px dt
  ∞ ∞ ∞
E Kx2 = k 2 k px qx+k =
P P P
(2k − 1) k px = 2 k k px − ex
k=1 k=1 k=1
h i
E min (Tx , n)
2 Rn Rn
= 0
t2 t px µx+t dt + n2 n px = 0
2tt px dt
h i n−1 n
E min (Kx , n) =
2 P 2
k k px qx+k + n2 n px =
P
(2k − 1) k px
k=1 k=1

Variance:
  2
V ar (Tx ) = E Tx2 − E [Tx ]
  2
V ar (Kx ) = E Kx2 − E [Kx ]

Recursive formulas:
o o o
ex = ex:n + n px ex+n

ex = ex:n + n px ex+n → ex = px + px ex+1

ex:n = ex:m + m px ex+m:n−m for m < n

o o o

ex:n = ex:m + m px ex+m:n−m for m < n → ex:n = px + px ex+1:n−1

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 15

C6. MORTALITY LAWS

 
Bcx ct −1

Gompertz’s law: µx = Bc for c > 1

x
→ t px =e − log c

 
Bcx ct −1

Makeham’s law: µx = A + Bc for c > 1 x

→ t px =e −At− log c

 
k (x+t)n+1 −xn+1
Weibull distribution: µx = kxn → t px = e− n+1

Exponential distribution: µx = µ → t px = e−µt

(Constant force of mortality)

Uniform distribution: µx = 1
ω−x for 0 ≤ x ≤ ω → t px =1− t
ω−x
 α
Beta distribution: µx = α
ω−x for 0 ≤ x ≤ ω → t px = 1− t
ω−x

C7. APPROXIMATIONS

Sqx
UDD between integral ages: lx+s = lx − sdx → s qx = sq x s qx+t = qx = s px µx+s
1 − tqx

CFM between integral ages: µx+s = − log px

s s s
lx+s = lx × (px ) → s px = (px ) s px+t = (px )

These are for 0 ≤ s, t ≤ 1 and 0 ≤ s + t ≤ 1.

C8. SELECT SURVIVAL MODEL

k-year select period: q[x]+h < qx+h for h < k

q[x]+h = qx+h for h ≥ k

p[x]+h > px+h for h < k

p[x]+h = px+h for h ≥ k

D. INSURANCE
D1. ACTUARIAL FUNCTIONS

Endowment:
2
n Ex =A 1 = v n n px 2
n Ex = 2A 1 = (v n ) n px
x: n x: n

Insurance (Continuous):
R∞ R∞ 2
Āx = 0
v t t px µx+t dt 2
Āx = 0
(v t ) t px µx+t dt
Rn Rn 2
Ā 1 = 0
v t t px µx+t dt 2
Ā 1 = 0
(v t ) t px µx+t dt
x:n x:n
Rn Rn 2 2
Āx:n = 0
v t t px µx+t dt + v n n px 2
Āx:n = 0
(v t ) t px µx+t dt + (v n ) n px

∞ ∞
Insurance (Discrete):

2
v k+1 k px qx+k 2
v k+1
P P
Ax = Ax = k px qx+k
k=0 k=0

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 16

n−1 n−1
2
v k+1 k px qx+k 2
v k+1
P P
A1 = A1 = k px qx+k
x:n x:n
k=0 k=0
n−1 n−1
2 2
v k+1 k px qx+k + v n n px 2
v k+1 qx+k + (v n )
P P
Ax:n = Ax:n = k px n px
k=0 k=0
nm−1
Insurance (mthly):
(m)
v k/m+1/m k/m px
P
A1 = 1/m qx+k/m
x:n k=0

Relations: Āx = Ā 1 + n Ex Āx+n Ax = A 1 + n Ex Ax+n

x:n x:n

Āx:n = Ā 1 + n Ex Ax:n = A 1 + n Ex
x:n x:n

n| Āx = n Ex Āx+n n| Ax = n Ex Ax+n

Recursive formulas: Ax = vqx + vpx Ax+1 2

Ax = v 2 qx + v 2 px 2 Ax+1

2
A1 = vqx + vpx A 1 A1 = v 2 qx + v 2 px 2 A 1
x:n x+1:n−1 x:n x+1:n−1

D2. PRESENT VALUES

Death benefit paid Death benefit paid

Policy at the moment of death at the end of the year of death

Whole life insurance Z = bv Tx , Tx > 0 Z = bv Kx +1 , Kx = 0, 1, 2, . . . , ∞

 
bv Tx , Tx < n
 bv Kx +1 , Kx = 0, 1, 2, . . . , n − 1

n-year term insurance Z= Z=
0, Tx ≥ n 0, Kx = n, n + 1, . . . , ∞
 

 
bv Tx ,

Tx < n bv Kx +1 ,

Kx = 0, 1, 2, . . . , n − 1
n-year endowment insurance Z= Z=
bv n ,

Tx ≥ n bv n ,

Kx = n, n + 1, . . . , ∞

0,

Tx < n
n-year pure endowment Z=
bv n ,

Tx ≥ n
 
0, Tx < n 0, Kx = 0, 1, 2, . . . , n − 1
 
n-year deferred whole life insurance Z= Z=
bv Tx ,

Tx ≥ n bv Kx +1 ,

Kx = n, n + 1, . . . , ∞

D3. EXPECTED PRESENT VALUES

Death benefit paid Death benefit paid

Policy at the moment of death at the end of the year of death

Whole life insurance E [Z] = bĀx E [Z] = bAx

n-year term insurance E [Z] = bĀ 1 E [Z] = bA 1

x:n x:n

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 17

n-year endowment insurance E [Z] = bĀx:n E [Z] = bAx:n

n-year pure endowment E [Z] = bn Ex

n-year deferred whole life insurance E [Z] = bn| Āx E [Z] = bn| Ax

D4. VARIANCE OF PRESENT VALUES

Death benefit paid Death benefit paid

Policy at the moment of death at the end of the year of death

2   
Whole life insurance
2
V ar (Z) = b2 2
Āx − Āx V ar (Z) = b2 2
Ax − (Ax )
  2    2 
n-year term insurance V ar (Z) = b 2 2
Ā 1 − Ā 1 V ar (Z) = b 2 2
A1 − A1
x:n x:n x:n x:n

2  
2 
n-year endowment insurance V ar (Z) = b2 2
Āx:n − Āx:n V ar (Z) = b2 2
Ax:n − Ax:n
 
n-year pure endowment
2
V ar (Z) = b2 2
nEx − (n Ex )

2  
2 
n-year deferred whole life insurance V ar (Z) = b2 2
n| Āx − n| Āx V ar (Z) = b2 2
n| Ax − n| Ax

D5. APPROXIMATIONS

UDD between integral ages: Āx = δi Ax Ā 1 = δi A 1 Āx:n 6= δi Ax:n .

x:n x:n

(m) i
Ax = A
i(m) x

2 2i+i2 2
Āx = 2δ Ax

Claims acceleration approach: Āx = (1 + i)0.5 Ax

0.5 0.5
Ā 1 = (1 + i) A1 Āx:n 6= (1 + i) Ax:n .
x:n x:n
m−1
(m)
Ax = (1 + i) 2m
Ax

2
Āx = (1 + i) 2 Ax

E. ANNUITIES
E1. ACTUARIAL FUNCTIONS
R ∞  1−vt  R ∞  1−v2t 
Annuity (Continuous): āx = 0 δ t px µx+t dt 2
āx = 0 2δ t px µx+t dt
R n  1−vt  1−v n

2
R n  1−v2t  
1−v 2n

āx:n = 0 δ t px µx+t dt + δ n px āx:n = 0 2δ t px µx+t dt + 2δ n px

∞   ∞  
Annuity (Discrete): 1−v k+1 2 1−v 2k+2
P P
äx = d k px qx+k äx = 2d−d2 k px qx+k
k=0 k=0
n−1
P  n−1
P  1−v2k+2   
1−v k+1 1−v n 1−v 2n

2
äx:n = d k px qx+k + d n px äx:n = 2d−d2 k px qx+k + 2d−d2 n px
k=0 k=0

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 18

Simpler formulas:
R∞ R∞
āx = 0
v t t px dt 2
āx = 0
v 2t t px dt
Rn Rn
āx:n = 0
v t t px dt 2
āx:n = 0
v 2t t px dt
∞ ∞
v k k px 2
v 2k k px
P P
äx = äx =
k=0 k=0
n−1 n−1
v k k px 2
v 2k k px
P P
äx:n = äx:n =
k=0 k=0
nm−1
(m) P 1 k/m
äx:n = mv k/m px
k=0

Relations: āx = āx:n + n Ex āx+n äx = äx:n + n Ex äx+n

n| āx = n Ex āx+n n| äx = n Ex äx+n

āx:n = ā n + n Ex āx+n äx:n = ä n + n Ex äx+n

äx = 1 + ax

äx:n = 1 + ax:n−1 = 1 + ax:n − n Ex

Recursive formulas: äx = 1 + vpx äx+1 2

äx = 1 + v 2 px 2 äx+1

2
äx:n = 1 + vpx äx+1:n−1 äx:n = 1 + v 2 px 2 äx+1:n−1

Insurance to annuity: āx = 1−Āx

δ
2
āx = 1−2 Āx
2δ

1−Āx:n 2 1−2 Āx:n

āx:n = δ āx:n = 2δ

1−Ax 2 1−2 Ax
äx = d äx = 2d−d2

1−Ax:n 2 1−2 Ax:n

äx:n = d äx:n = 2d−d2
(m)
(m) 1−Ax:n
äx:n = d(m)

E2. PRESENT VALUES

Payments made Payments at the

Policy continuously beginning of each year
   
Whole life annuity Y =b 1−v Tx
δ , Tx > 0 Y =b 1−v Kx +1
d , Kx = 0, 1, 2, . . . , ∞
     
b 1−vT x ,

Tx < n b 1−vKx +1 , Kx = 0, 1, 2, . . . , n − 1

n-year term annuity
δ d
Y = Y =
n
b
 1−v


, Tx ≥ n b 1−vn
,

Kx = n, n + 1, . . . , ∞
δ d

 
1−v n 1−v n



b , Tx < n b , Kx = 0, 1, 2, . . . , n − 1
 
n-year certain whole life annuity
δ d
Y =   Y =  
b 1−vTx ,

Tx ≥ n b 1−vKx +1 , Kx = n, n + 1, . . . , ∞

δ d

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 19

 
0, Tx < n 0, Kx = 0, 1, 2, . . . , n − 1
 
n-year deferred whole life annuity Y =   Y =  
n T n K +1
b v −v x , Tx ≥ n
 b v −v x

, Kx = n, n + 1, . . . , ∞
δ d

E3. EXPECTED PRESENT VALUES

Payments made Payments at the

Policy continuously beginning of each year

Whole life annuity E [Y ] = bāx E [Y ] = bäx

n-year term annuity E [Y ] = bāx:n E [Y ] = bäx:n

n-year certain whole life annuity E [Y ] = bāx:n E [Y ] = bäx:n

n-year deferred whole life annuity E [Y ] = bn| āx E [Y ] = bn| äx

E4. VARIANCE OF PRESENT VALUES

Payments made Payments at the

Policy continuously beginning of each year
2

2
Whole life annuity
2
Āx − Āx Ax −(Ax )2
V ar (Y ) = b2 δ2 V ar (Y ) = b2 d2

2

2 2

2
n-year term annuity
Āx:n − Āx:n Ax:n − Ax:n
V ar (Y ) = b2 δ2 V ar (Y ) = b2 d2

E5. APPROXIMATIONS

UDD between integral ages: āx = id

δ 2 äx − i−δ
δ2

(m) id i−i(m)
äx = ä
i(m) d(m) x
− i(m) d(m)

(m) id i−i(m)
äx:n = ä
i(m) d(m) x:n
− i(m) d(m)
(1 − n Ex )

Woolhouse formula, 2 terms: āx ≈ äx − 1

2

(m) m−1
äx ≈ äx − 2m

(m) m−1
äx:n ≈ äx:n − 2m (1 − n Ex )

Woolhouse formula, 3 terms: āx ≈ äx − 1

2 − 1
12 (µx + δ)

m2 −1
where µx ≈ − 12 (log px−1 + log px ) if not given.
(m) m−1
äx ≈ äx − 2m − 12m2 (µx + δ)

(m) m−1 m2 −1
äx:n ≈ äx:n − 2m (1 − n Ex ) − 12m2 (µx + δ − n Ex (µx+n + δ))

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 20

F. PREMIUMS
F1. ACTUARIAL FUNCTIONS
A1
Premium (Discrete): Ax x:n Ax:n
Px = äx P1 = äx:n Px:n = äx:n
x:n
  Ā 1
Premium (Continuous):

Āx x:n

Āx:n
P̄ Āx = āx P̄ Ā 1 = āx:n P̄ Āx:n = āx:n
x:n

F2. EQUIVALENCE PRINCIPLE

Loss-at-issue: 0L
n
= P V (Benefits) − P V (Premiums)

0L
g
= P V (Benefits) + P V (Expenses) − P V (Premiums)

Equivalence principle: E[0 L] = EP V (Benefits) − EP V (Premiums) = 0

E[0 Lg ] = EP V (Benefits) + EP V (Expenses) − EP V (Premiums) = 0

F3. EXPECTATION AND VARIANCE

Fully continuous whole life insurance Fully discrete whole life insurance
   
g 1−v Tx g 1−v Kx +1
0L = (b + E) v Tx − (G − e) δ , Tx ≥ 0 0L = (b + E) v Kx +1 − (G − e) d , Kx = 0, 1, 2, . . . , ∞

g G−e G−e g G−e G−e

0L = b+E+ δ Z− δ 0L = b+E+ d Z− d

E [0 Lg ] = (b + E) Āx − (G − e) āx E [0 Lg ] = (b + E) Ax − (G − e) äx


2   
G−e 2 G−e 2 2
V ar (0 Lg ) = b + E + V ar (0 Lg ) = b + E +

2

2
δ Āx − Āx d Ax − (Ax )

Fully continuous n-year endowment insurance Fully discrete n-year endowment insurance
   
(b + E) v Tx − (G − e) 1−vTx , Tx < n
 
 (b + E) v Kx +1 − (G − e) 1−vKx +1 , Kx = 0, 1, . . . , n − 1

g δ g d
0L = 0L =
(b + E) v n − (G − e) 1−vn
,

Tx ≥ n (b + E) v n − (G − e) 1−vn
,

Kx = n, n + 1, . . . , ∞
δ d

g G−e G−e g G−e G−e

0L = b+E+ δ Z− δ 0L = b+E+ d Z− d

E [0 Lg ] = (b + E) Āx:n − (G − e) āx:n E [0 Lg ] = (b + E) Ax:n − (G − e) äx:n


2  
2 
G−e 2 G−e 2
V ar (0 Lg ) = b + E + V ar (0 Lg ) = b + E +

2

2
δ Āx:n − Āx:n d Ax:n − Ax:n

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 21

F4. PORTFOLIO PERCENTILE PRINCIPLE

Aggregate losses: S = Lg1 + Lg2 + · · · + Lgn ∼N

˙ µ, σ 2

Mean: µ = E[S] = nE [0 Lg ]

Variance: σ 2 = Var(S) = n Var (0 Lg )

 
Find premium such that: Pr(S < 0) ≈ N 0−nE[0 Lg ]
p
n Var(0 Lg )
=α → 0−nE[0 Lg ]
p
n Var(0 Lg )
= zα

G. RESERVES
G1. PROSPECTIVE FORMULA

Net premium reserve: tV

n
= E [t L | Tx ≥ t] = EP Vt (Benefits) − EP Vt (Net Premiums)

Gross premium reserve: tV

g
= E [t Lg | Tx ≥ t] = EP Vt (Benefits) + EP Vt (Expenses)
− EP Vt (Gross Premiums)

Expense reserve: tV
e
= t V g − t V n = EP Vt (Expenses)−EP Vt (Expense Loadings)

G2. EXPECTATION AND VARIANCE

Fully continuous whole life insurance Fully discrete whole life insurance
   
g 1−v Tx+t g 1−v Kx+k +1
tL = (b + E) v Tx+t − (G − e) δ , Tx+t ≥ 0 kL = (b + E) v Kx+k +1 − (G − e) d ,

Kx+k = 0, 1, 2, . . . , ∞

g G−e G−e g G−e G−e

tL = b+E+ δ Z− δ kL = b+E+ d Z− d

E [t Lg | Tx ≥ t] = (b + E)Āx+t − (G − e)āx+t E [k Lg | Kx ≥ k] = (b + E)Ax+k − (G − e)äx+k


2   
G−e 2 G−e 2 2



V ar (t Lg | Tx ≥ t) = b + E + δ
2
Āx+t − Āx+t V ar (k Lg | Kx ≥ k) = b + E + d
2
Ax+k − (Ax+k )

Fully continuous n-year endowment insurance Fully discrete n-year endowment insurance
  
Kx+k +1



 (b + E) v Kx+k +1 − (G − e) 1−v d ,
   

(b + E) v Tx+t − (G − e) 1−vTx+t , Tx+t < n − t

Kx+k = 0, . . . , n − k − 1
 

g δ g
tL =   kL =  
(b + E) v n−t − (G − e) 1−vn−t ,

Tx+t ≥ n − t

 (b + E) v n−k − (G − e) 1−vd
n−k
,
δ





Kx+k = n − k, n − k + 1, . . .



g G−e G−e g G−e G−e

tL = b+E+ δ Z− δ kL = b+E+ d Z− d

E [t Lg | Tx ≥ t] = (b + E)Āx+t:n−t − (G − e)āx+t:n−t E [k Lg | Kx ≥ k] = (b + E)Ax+k:n−k − (G − e)äx+k:n−k

  2    2 

2 2
2 2
V ar (t Lg | Tx ≥ t) = b + E + G−eδ Āx+t:n−t − Āx+t:n−t V ar (k Lg | Kx ≥ k) = b + E + G−e
d Ax+k:n−k − Ax+k:n−k

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com

 ACTEX Learning Exam FAM Page 22

G3. RECURSIVE FORMULA

Net premium reserve: (k V + P ) (1 + i) = bqx+k + k+1 V px+k

Gross premium Reserve: (k V g + G − e) (1 + i) = (b + E) qx+k + k+1 V g px+k

For reserves between premium dates: where 0 < s < 1.

s
(k V + P ) (1 + ik ) = b v 1−s s qx+k + k+s V s px+k

where 0 < s < 1.

1−s
k+s V (1 + ik ) =b 1−s qx+k+s + k+1 V 1−s px+k+s

For a special policy that pays (k−1 V + P ) (1 + i) = (F A + k V ) qx+k−1 + k V px+k−1

F A + k V upon death:

Rearrange: kV = (k−1 V + P ) (1 + i) − F A qx+k−1

k
Obtain the formula: = P ä k (1 + i)k − F A qx+j−1 (1 + i)k−j
P
kV
j=1

G4. FPT RESERVES

First year premium: FPT

 
E 0L = vqx − α = 0 → α = vqx

Renewal year premium: 1V

FPT
= Ax+1 − βäx+1 = 0 → β= Ax+1
äx+1 whole life
A 1

1V
FPT
=A 1 − βäx+1:n−1 = 0 → β= x+1:n−1
äx+1:n−1 n-year term
x+1:n−1

Ax+1:n−1
1V
FPT
= Ax+1:n−1 − βäx+1:n−1 = 0 → β= äx+1:n−1 n-year endowment

www.ACTEXLearning.com www.actuarialbookstore.com www.studymanuals.com