## Brownian motion via Hilbert basis

Brownian motion is a central object of probability theory. One perspective to look at its construction uses Hilbert spaces.
Let be a Hilbert basis equipped with standard scalar product . Hence, every can be written in a unique way as

where and . Let be an i.i.d. sequeence of standard gaussian random variables and consider

Using the independence of we obtain

Brownian motion is then given by (a continuous modification of) , . Hence, “all” we need to do is to find a Hilbert basis. The most known is the following.

#### Haar basis

Here we refer to http://datatreker.com/haar-basis for an introduction.

haar_mother <- function(x){
(x >0 & x <= 0.5) - (x >0.5 & x <= 1)
}
haar <- function(x,j,k){
(2^(j/2) * haar_mother(2^j*x-k))
}
set.seed(42)

j_max <- 10
n_max <- 11  # maximal resolution
delta <- 2^{-n_max}
x <- (0: 2^n_max)/2^n_max
xi_0 <- rnorm(1) # xi corresponding to constant function
xi <- list()     # list of random variables xi
for (j in 0:j_max){
xi[[j+1]] <- rep(0, 2^j)
for (k  in 0:(2^j-1))
{
xi[[j+1]][k+1] <-rnorm(1)
}
}

# data.frame containing approximation of the Brownian motion
df <- data.frame(x=numeric(),
y=numeric(),
id=numeric())
for (i in x){
y <- c(1,rep(1, i*2^n_max), rep(0, 2^n_max-i*2^n_max))
alpha <- list() # wavelet coefficients for y
bm <- sum(y)/(length(y)+1) * xi_0
for (j in 0:j_max){
alpha[[j+1]] <- rep(0, 2^j)
for (k  in 0:(2^j-1))
{
alpha[[j+1]][k+1] <-sum(haar(x,j,k)*y)*delta
bm <- bm + alpha[[j+1]][k+1]*xi[[j+1]][k+1]
}
df_new <- data.frame(x=i,
y=bm,
id=j)
df <- bind_rows(df, df_new)
}
}

#### Animation using gganimate

ggplot(df, aes(x=x, y=y)) +
geom_line()+
transition_states(
id,
transition_length = 2,
state_length = 1
) +
labs(title = "Construction of Brownian motion using Haar basis. Step: {closest_state}") +
ease_aes('sine-in-out')

The result looks the same as with Levy’s construction which is not astonishing. The implementation above is not optimal and is much slower than the previous one.

It might be worth to consider a different Hilbert basis.

#### Trigonometric Hilbert basis

The trigonometric basis is defined as

together with .

set.seed(42)
N_max <-  200 # number of trigonometric basis functions used
n_max <- 11  # maximal resolution
delta <- 2^{-n_max}
x <- (0: 2^n_max)/2^n_max
xi <- rnorm(N_max+1)     # list of random variables xi

# data.frame containing approximation of the Brownian motion
df <- data.frame(x=numeric(),
y=numeric(),
id=numeric())

for (i in x){
y <- c(1,rep(1, i*2^n_max), rep(0, 2^n_max-i*2^n_max))
alpha <- numeric() # wavelet coefficients for y
alpha[1] <- mean(y)
bm <-  alpha[1] * xi[1]
for (n in 1:N_max){
alpha[n] <-sum(sqrt(2)*cos(x*pi*n)*y)*delta
bm <- bm + alpha[n]*xi[n]
df_new <- data.frame(x=i,
y=bm,
id=n)
df <- bind_rows(df, df_new)}
}

#### Animation using gganimate

p <- ggplot(df, aes(x=x, y=y)) +
geom_line()+
transition_states(
id,
transition_length = 2,
state_length = 1
) +
labs(title = "Construction of Brownian motion using trigonometric basis. Step: {closest_state}") +
ease_aes('sine-in-out')
animate(p, nframes=1000)

The above convergence is very slow. The irregularity properties of the Brownian motion are not yet very visible.

## Haar basis

Hilbert spaces play a prominent role in various fields of mathematics. An orthonormal basis of such a space is called a Hilbert basis. The purpose of this blog is to illustrate a very clasic and basic Hilbert basis – the Haar basis.

Let be a Hilbert basis of equipped with the standard scalar product . Hence, every can be written in a unique way as

where and .

A classic Hilbert basis consists of Haar functions that are supported on . They are defined using the Haar wavelet:

The Haar basis consists then of rescaled (by ) versions of shifted by ,

together with the constant function . (Note: the constant function has to be added since we consider the interval and not ). In R this looks like:

haar_mother <- function(x){
(x >0 & x <= 0.5) - (x >0.5 & x <= 1)
}
haar <- function(x,j,k){
2^(j/2) * haar_mother(2^j*x-k)
}

#### Animation of the Haar basis

j_max <- 3    # maximal depth
n_max <- 10   # resolution of grid
df <- data.frame(x=numeric(),
y=numeric(),
id=numeric())

x <- (1: 2^n_max)/2^n_max
id<-1
for (j in 0:j_max){
for (k  in 0:(2^j-1))
{
df_new<- data.frame(x=x,
y=haar(x,j,k),
id=id)
df <- bind_rows(df, df_new)
id <- id+1
}
}

ggplot(df, aes(x=x, y=y)) +
geom_step()+
transition_states(
id,
transition_length = 2,
state_length = 5
) +
labs(title = 'Illustration of Haar basis') +
ease_aes('sine-in-out')

#### Approximation via Haar basis

Now, every function can be written as

The coefficients are called the Haar Wavelet coefficients. Let us calculate them in the discrete setting. We give us a mesh and values and some function :

j_max <- 12
n_max <- 13  # maximal resolution
delta <- 2^{-n_max}
x <- (1: 2^n_max)/2^n_max
y <- x* sin(1/x)
alpha <- list()  # list of Haar coefficients
for (j in 0:j_max){
alpha[[j+1]] <- rep(0, 2^j)
for (k  in 0:(2^j-1))
{
alpha[[j+1]][k+1] <-sum(haar(x,j,k)*y)*delta  # approximation of scalar product
}

}

y_approx <- rep(0, length(y))  # approximated values of y

#### Calculating the approximations for different values of j

df <- data.frame(x=numeric(),
y=numeric(),
id=numeric())

y_approx <-  mean(y) # this is the contribution of the constant function
for (j in 0:j_max){
for (k  in 0:(2^j-1))
{
y_approx <- y_approx + alpha[[j+1]][k+1]*haar(x,j,k)
}
df_new <- data.frame(x=x,
y=y_approx,
id=j)
df <- bind_rows(df, df_new)
}

## Animation of convergence


ggplot(df, aes(x=x, y=y)) +
geom_step()+
transition_states(
id,
transition_length = 2,
state_length = 1
) +
labs(title = "Haar basis approximation of x sin(1/x). Depth:  {closest_state}") +
ease_aes('sine-in-out')

## Lévy’s construction of Brownian motion

Brownian motion is a central object of probability theory. The idea of Lévy’s construction is to construct the Brownian motion step by step on finite sets

of dyadic points. As is dense in the Brownian motion is then obtained as the uniform limit of linear interpolation on .
It is pretty easy to illustrate this construction using R and the package gganimate. We use the same notation as in the proof of Wiener’s theorem given on page 23 in “Brownian motion” by Peter Mörters and Yuval Peres.

library(dplyr)
library(gganimate)
library(transformr)
set.seed(42)
n_max <- 14                # maximal number of steps
D <- (0: 2^n_max)/2^n_max   # this is in fact D_n_max
B <- list()
Z <- rnorm(1)
B[[1]] <- c(0, Z/2 + rnorm(1), Z)
for (n in 2:n_max){
B[[n]] <- rep(0, 2^n+1)
index_known <- seq(1,2^n+1, by=2) # indices where values are known from previous steps
B[[n]][index_known] <- B[[n-1]]
index_unknown <- seq(2, 2^n, by=2) # indices where values are not yet defined
for (d in index_unknown){
B[[n]][d] <- 0.5*(B[[n]][d-1]+B[[n]][d+1])+ rnorm(1)*2^(-(n+1)/2)
}
}

## interpolation and transformation in a data.frame
df <- data.frame(time=numeric(),
value=numeric(),
id=numeric())

for (n in 1:n_max){
D_n<-(0: 2^n)/2^n
B_interpol<- approx(D_n, B[[n]], xout = D)\$y   # interpolation
df_new <- data.frame(time=D,
value=B_interpol,
id=n)
df <- bind_rows(df, df_new)
}

## animation
ggplot(df, aes(x=time, y=value)) +
geom_line()+
transition_states(
id,
transition_length = 2,
state_length = 1
) +
labs(title = 'Levys construction of Brownian motion.  Step: {closest_state}', x = 'time', y = 'position') +
ease_aes('sine-in-out')`