Part 4 R and random variables

Dr. Nguyen Quang Huy

March 06, 2020

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 1 / 43
Where does “randomness” come from ?

Why using “sample(0:1,1)” can generate a random number X where

P (X = 0) = P (X = 1) = 0.5

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 2 / 43
Where does “randomness” come from ?

Why using “sample(0:1,1)” can generate a random number X where

P (X = 0) = P (X = 1) = 0.5

RANDOM
NUMBER
GENERATOR

True random Pseudo-random

generator generator

Human sources Natural sources

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 2 / 43
True random generator - Human sources

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 3 / 43
True random generator - Natural source
Ratioactive decay:

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 4 / 43
True random generator

Pros: They are truly random numbers.

Cons:
Random numbers are generated slowly (compared to Pseudo-radom
generator)
Need a huge storage to record.
Systematic error.

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 5 / 43
Pseudo-random generator (Computer algorithms)
Pseudo-random number generator (PRNG) are computer algorithms to
generate sequences of Uniform[0,1] random variables u1 , u2 , · · ·, ud .
PRNG is defined by {S, s1 , f , g} where
S is state universe (all possible states).
s1 ∈ S is the seed.
f : S → S.
g : S → [0, 1].
Uniform random variable u1 , u2 , · · · , ud are simulated as follows:
u1 = g(s1 ).
sk = f (sk−1 ).
uk = g(sk ).
*Note: For any PRNG, the number
Dr. Nguyen Quang Huy
of state d, d = number March
Part 4 R and random variables
of elements
06, 2020
in
6 / 43
Pseudo-random generator

SEED

f f f f f
s1 s2 s3 ··· sd s1

g g g g g

u1 u2 u3 ··· ud u1

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 7 / 43
Pseudo-random generator - linear congruence
s1 is an integer from 1 to M
f (s) = (A × s + C ) mod M i.e. sk+1 = (A × sk + C ) mod M.
s sk
g(s) = i.e. uk =
M M
Let M = 231 − 1; A = 75 ; C = 0 and s1 = 10.
Generate u1 , u2 , · · · , uN from above algorithm (N = 106 ) ?
Check the correlation between uk−1 and uk ?
Calculate from the sample the probability pi = P(U ≤ vi ) with
!
1 2 k
v= , ,··· ,
k +1 k +1 k +1
and plot pi against v ?
Calculate d - smallest integer number such that ud+1 = u1 (from 10 to
Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 8 / 43
Wichmann Hill algorithm
Let p1 = 30269; p2 = 30307 and p3 = 30323.
Seed s1 = (s1,1 , s1,2 , s1,3 ) where s1,j are integer numbers from 1 to
30,000.
Function f :
sk,1 = 171 × s(k−1),1 mod p1
sk,2 = 172 × s(k−1),2 mod p2
sk,3 = 170 × s(k−1),3 mod p3

Function g: uk = vk − [vk ] where [vk ] is integer part of vk and

sk,1 sk,2 sk,3
vk = + +
p1 p2 p3

*Note: Wichmann Hill algorithm’s cycle is ∼ 6.95 × 1012 .

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 9 / 43
Wichmann Hill algorithm

With s1 = (100, 200, 300), generate u1 , u2 , · · · , uN from above

algorithm (N = 106 ) ?
Check the correlation between uk−1 and uk ?
Calculate from the sample the probability pi = P(U ≤ vi ) with
!
1 2 k
v= , ,··· ,
k +1 k +1 k +1

and plot pi against v ?

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 10 / 43
Mersenne-Twister algorithm

Default PRNG in almost all softwares: R, Excel, MATLAB, Python, C++,

Mathematica . . .
Developed in 1997 by Makoto Matsumoto and Takuji Nishimura
Period length is chosen to be a Mersenne prime
Maximum period is 219937 − 1 ≈ 105981
"Mersenne twister: a 623-dimensionally equidistributed uniform
pseudo-random number generator"
Generate an uniform vector of length N in [0, 1] with R: runif (N, 0, 1)

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 11 / 43
Generate continuous random variables

If X is continuous random variable with distribution function

F (x ) = P(X ≤ x ), the random variable U defined by U = F (X) is uniformly
distributed on [0, 1].
→ to simulate X with distribution F :
1. Simulate a random number U in [0, 1].
2. Return X = F ← (U) where F ← is the inverse function of F (Using
"uniroot" function in R to solve it if there is no analytical form of F ← )
3. Repeat above steps n times to has n realization of X.

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 12 / 43
Generate continuous random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 13 / 43
Generate continuous random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 14 / 43
Generate continuous random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 15 / 43
Generate continuous random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 16 / 43
Generate continuous random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 17 / 43
Generate continuous random variables

Example 1: A continuous random variable X ≥ 0, has distribution function:

1 2
P(X ≤ x ) = 1 − −
3(x + 1) 3(x + 1)2

Is there a close form solution for F ← (X ) = U. ?

Simulate a sample of n = 106 of value of X.
Calculate from the sample: P(X ≤ x ) and compare them with the true
probabilities.
(If there is not a closed-form for inverse function F ← we can use uniroot
function in R. "Unifoot(f , c(a, b)) $root” returns the root r of equation
f (x ) = 0 where a < r < b. Note that f (a) and f (b) must have opposite
sign).

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 18 / 43
Generate continuous random variables

1 t 2t 2
Example 1: Closed-form solution: Let = t then 1 − − = u.
x +1 3 3
p
2 1 + 24(1 − u) − 1
→ 2t + t − 3(1 − u) = 0 → t =
4
4
→ X=p −1
1 + 24(1 − U) − 1

F.inverse<-function(u){
F.inverse<-4/(sqrt(1+24*(1-u))-1)-1}
U<-runif(10ˆ6,0,1)
X<-F.inverse(U)

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 19 / 43
Generate continuous random variables

Example 2: The total expenses of company ABC is calculated by the sales

amount X as follow:

Total expense = x 0.1 + x 0.2 + x 0.3

where X is a random variable with distribution function

P(X ≤ x ) = 1 − 0.3e −x − 0.7e −1.5x

What is the expectation of total expense ? What is VaR90% of total expense

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 20 / 43
Generate continuous random variables
Example 2:
n<-10ˆ4
u<-runif(n,0,1)
x<-rep(0,n)
for (i in 1:n){
F1<-function(x){F1<-1-0.3*exp(-0.5*x)-0.7*exp(-1.5*x)-u[i]}
x[i]<-uniroot(F1,c(0,1000))$root}
totalexpense<-xˆ0.1+xˆ0.2+xˆ0.3
mean(totalexpense)

## [1] 2.735214
quantile(totalexpense,0.9)

## 90%
## 3.610101
Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 21 / 43
Generate continuous random variables

Variable Density Generate n obsers

1
Uniform(a, b) runif (n, a, b)
b−a
Γ(a + b) a−1
Beta(a, b) x (1 − x )b−1 rbeta(n, a, b)
Γ(a) Γ(b)
Exponential(β) βe −βx rexp(n, β)
β α α−1 −βx
Gamma(α, β) x e rgamma(n, α, β)
Γ(α)
(x − µ)2
!
1
Normal (µ, σ) √ exp − rnorm(n, µ, σ)
2πσ 2σ 2
Log-normal (µ, σ) Ln(X ) ∼ Normal(µ, σ) rlnorm(n, µ, σ)

Pareto(α, β) α β α (x + β)−α−1 NA

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 22 / 43
Generate discrete random variables

Example 1: Let X be a Bernoulli random variable where

P(X = 1) = p
P(X = 0) = 1 − p

To simulate X
1. Generate U ∼ Uniform(0,1).
2. If U ≤ (1 − p) return 0; return 1 otherwise.
p<-0.3
u<-runif(1,0,1)
x<-ifelse(u<=(1-p),0,1)
print(x)

## [1] 0
Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 23 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 24 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 25 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 26 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 27 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 28 / 43
Generate discrete random variables

Example 1: Generate a vector contains n Bernoulli random variables

p<-0.3
n<-10ˆ4
u<-runif(n,0,1)
x<-ifelse(u<=(1-p),0,1)
print(x[1:10])

## [1] 0 0 0 0 0 0 0 0 1 1
sum(x)/n #should be close to p

## [1] 0.2949

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 29 / 43
Generate discrete random variables

Discrete random variable X has probability mass function

p0 = P(X = 0)
p1 = P(X = 1)
···
pk = P(X = k)
···
k
Let qk = P(X ≤ k) = pi with k = 0, 1, 2, · · ·; to simulate X :
P
i=0

1. Generate an uniform random variable U in [0, 1].

2. Find k such that qk−1 ≤ U < qk and return k.

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 30 / 43
Generate discrete random variables

∞
P
with pi = 1; and
i=0
k
qk = P(X ≤ k) =
P
pi
i=0

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 31 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 32 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 33 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 34 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 35 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 36 / 43
Generate discrete random variables

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 37 / 43
Generate discrete random variables

Example 2 A discrete random variable X , X ≥ 0, has probability mass

function:
1 2 1
P(X = k) = p (1 − p)k + e −1
3 3 k!
where p = 0.2
Simulate a sample of N = 106 of value of X .
Calculate from the sample: P(X = 0), P(X = 1), · · · , and compare
them with the true probabilities.
WARNING: vector indices in R start with 1

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 38 / 43
Generate discrete random variables
n<-100
p<-0.2
P0<-1/3*p+2/3*exp(-1)
P<-rep(0,n)
for (i in 1:n){P[i]<-1/3*p*(1-p)ˆi+2/3*exp(-1)/factorial(i)}
N<-10ˆ6
X<-rep(0,N)
U<-runif(N,0,1)
for (i in 1:N){
if (U[i]<P0){k<-0} else {
for (k in 1:n){if (P0+sum(P[1:k])>U[i]) {break}}}
X[i]<-k}
sum(X==2)/N

## [1] 0.165482

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 39 / 43
Generate discrete random variables

Variables P(X = k) Generate N obsers

Bernoulli(p) p k (1 − p)(1−k) rbinom(N, 1, p)
n
k
Binomial(n, p) k
p (1 − p)(n−k) rbinom(N, n, p)
Geometric(p) p (1 − p)k rnbinom(N, 1, p)
r +k−1
r
Negative binomial(r , p) k
p (1 − p)k rnbinom(N, r , p)
λk
Poisson (λ) e −λ rpois(N, λ)
k!

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 40 / 43
Simulation exercises

Exercise 1: A random variable X has cumulative distribution function:




 0 x <0
 x2

0≤x <1

F (x ) = 4
x +1


 4 1≤x <2

1 2≤x



2
Calculate E(e X + e X ).

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 41 / 43
Simulation exercises

Exercise 2: Given that X1 , X2 , · · · , X5 are 5 independent Pareto random

variables of parameter:

α1 = 1.2; β1 = 2.0
α2 = 1.3; β2 = 1.8
α3 = 1.4; β3 = 1.6
α4 = 1.5; β4 = 1.4
α5 = 1.6; β5 = 1.2

Let S5 = X1 + X2 + X3 + X4 + X5 . Calculate
P(S5 > 10); P(S5 > 103 ); P(S5 > 105 ).
VaR99% (S5 ) and CVaR99% (S5 ).

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 42 / 43
Simulation exercises

Exercise 3 (Collective model): Total claim amount of a car driver in a

year is modeled by 
0
 N=0
SN = N
Xi N ≥ 1
P


i=1
where N is number of claims in the year and Xi is the claim amount of the
i th accident. N and Xi are independent variables. The following assumptions
are given about the number of claim N and the claim amounts Xi :
N is a Poisson random variable of parameter λ = 2.
Xi are i.i.d Pareto random variables of parameter α = 1.5 and β = 2.5.
Calculate P(SN > 20) and VaR99% (SN )

Dr. Nguyen Quang Huy Part 4 R and random variables March 06, 2020 43 / 43