sampling Archives – H. Paul Keeler

Simulating Poisson random variables in Fortran

The hint’s in the title. I wrote a simple function in Fortran for simulating (or sampling) Poisson random variables. (More precisely, I should say that the function generates Poisson variates.) I used the simple direct method. This method is based on the exponential inter-arrival times of the Poisson (stochastic) process.

My code should not be used for large Poisson parameter values (larger than, say, 20 or 30), as the code may be too slow. Other methods exist for larger parameter values, which I’ve discussed previously.

I just use the standard Fortran function random_number for generating (pseudo-)random numbers. I am not an expert in Fortran, but my Poisson function seems to work fine. I wrote and ran a simple test that estimates the first and second moments, which should match for Poisson variables.

My Fortran code is very similar to the code that I wrote in C and C#, which is located here. You should be able to run it on this website or similar ones that can compile Fortran (95) code.

Code

On this page I’ve only included the code of the functions for generating uniform and Poisson variates. The rest of the code, including the test, is located here.

!Poisson function -- returns a single Poisson random variable
function funPoissonSingle(lambda) result(randPoisson)
real, intent(in) :: lambda !input
real :: exp_lambda !constant for terminating loop
real :: randUni !uniform variable
real :: prodUni !product of uniform variables
integer :: randPoisson !Poisson variable

exp_lambda= exp(-lambda) 

!initialize variables
randPoisson = -1;
prodUni = 1;
do while (prodUni > exp_lambda)
   randUni = funUniformSingle() !generate uniform variable
   prodUni = prodUni * randUni; !update product
   randPoisson = randPoisson + 1 !increase Poisson variable
end do
end function

!Uniform function -- returns a standard uniform random variable
function funUniformSingle() result(randUni)
real randUni;
call random_seed
call random_number(randUni)

end function

Simulating a Poisson point process on a sphere

In this post I’ll describe how to simulate or sample a homogeneous Poisson point process on the surface of a sphere. I have already simulated this point process on a rectangle, triangle disk, and circle.

Of course, by sphere, I mean the everyday object that is the surface of a three-dimensional ball, where this two-dimensional object is often denoted by $S^2$. Mathematically, this is a generalization from a Poisson point process on a circle, which is slightly simpler than randomly positioning points on a disk. I recommend reading those two posts first, as a lot of the material presented here builds off them.

I have not needed such a simulation in my own work, but I imagine there are many reasons why you would want to simulate a Poisson point process on a sphere. As some motivation, we can imagine these points on a sphere representing, say, meteorites or lightning hitting the Earth.

The generating the number of points is not difficult. The trick is positioning the points on the sphere in a uniform way. As is often the case, there are various ways to do this, and I recommend this post, which lists the main ones. I will use two methods that I consider the most natural and intuitive ones, namely using spherical coordinates and normal random variables, which is what I did in the post on the circle.

Incidentally, a simple modification allows you to scatter the points uniformly inside the sphere, but you would typically say ball in mathematics, giving a Poisson point process inside a ball; see below for details.

Steps

As always, simulating a Poisson point process requires two steps.

Number of points

The number of points of a Poisson point process on the surface of a sphere of radius $r>0$ is a Poisson random variable with mean $\lambda S_2$, where $S_2=4\pi r^2$ is the surface area of the sphere. (In this post I give some details for simulating or sampling Poisson random variables or, more accurately, variates.)

Locations of points

For any homogeneous Poisson point process, we need to position the points uniformly on the underlying space, which is in this case is the sphere. I will outline two different methods for positioning the points randomly and uniformly on a sphere.

Method 1: Spherical coordinates

The first method is based on spherical coordinates $(\rho, \theta,\phi)$, where the radial coordinate $\rho\geq 0$, and the angular coordinates $0 \leq \theta\leq 2\pi$ and $0\leq \phi \leq \pi$. The change of coordinates gives $x=\rho\sin(\theta)\cos(\phi)$, $y=\rho\sin(\theta)\sin(\phi)$, and $z=\rho\cos(\phi)$.

Now we use Proposition 1.1 in this paper. For each point, we generate two uniform variables $V$ and $\Theta$ on the respective intervals $(-1,1)$ and $(0,2\pi)$. Then we place the point with the Cartesian coordinates

$$X = r \sqrt{1-V^2} \cos\Theta, $$

$$Y = r \sqrt{1-V^2}\sin\Theta, $$

$$ Z=r V. $$

This method places a uniform point on a sphere with a radius $r$.

How does it work?

I’ll skip the precise details, but give some interpretation of this method. The random variable $\Phi := \arccos V$ is the $\phi$-coordinate of the uniform point, which implies $\sin \Phi=\sqrt{1-V^2}$, due to basic trigonometric identities. The area element in polar coordinates is $dA = \rho^2 \sin\phi d\phi d\theta $, which is constant with respect to $\theta$. After integrating with respect to $\phi$, we see that the random variable $V=\cos\Phi$ needs to be uniform (instead of $\Phi$) to ensure the point is uniformly located on the surface.

Method 2: Normal random variables

For each point, we generate three standard normal or Gaussian random variables, say, $W_x$, $W_y$, and $W_z$, which are independent of each other. (The term standard here means the normal random variables have mean $\mu =0$ and standard deviation $\sigma=1$.) The three random variables are the Cartesian components of the random point. We rescale the components by the Euclidean norm, then multiply by the radius $r$, giving

$$X=\frac{rW_x}{(W_x^2+W_y^2+W_z^2)^{1/2}},$$

$$Y=\frac{rW_y}{(W_x^2+W_y^2+W_z^2)^{1/2}},$$

$$Z=\frac{rW_z}{(W_x^2+W_y^2+W_z^2)^{1/2}}.$$

These are the Cartesian coordinates of a point uniformly scattered on a sphere with radius $r$ and a centre at the origin.

How does it work?

The procedure is somewhat like the Box-Muller transform in reverse. In the post on the circle setting, I gave an outline of the proof, which I recommend reading. The joint density of the normal random variables is from a multivariate normal distribution with zero correlation. This joint density is constant on the sphere, implying that the angle of the point $(W_x, W_y, W_z)$ will be uniformly distributed.

The vector formed from the normal variables $(W_x, W_y,W_z)$ is a random variable with a chi distribution. We can see that the vector from the origin to the point $(X,Y,Z)$ has length one, because we rescaled it with the Euclidean norm.

Plotting

Depending on your plotting software, the points may more resemble points on an ellipsoid than a sphere due to the different scaling of the x, y and z axes. To fix this in MATLAB, run the command: axis square. In Python, it’s not straightforward to do this, as it seems to lack an automatic function, so I have skipped it.

Results

I have presented some results produced by code written in MATLAB and Python. The blue points are the Poisson points on the sphere. I have used a surface plot (with clear faces) to illustrate the underling sphere in black.

MATLAB

Python

Note: The aspect ratio in 3-D Python plots tends to squash the sphere slightly, but it is a sphere.

Code

The code for all my posts is located online here. For this post, the code in MATLAB and Python is here. In Python I used the library mpl_toolkits for doing 3-D plots.

Poisson point process inside the sphere

Perhaps you want to simulate a Poisson point process inside the ball. There are different ways we can do this, but I will describe just one way, which builds off Method 1 for positioning the points uniformly. (In a later post, I will modify Method 2, giving a way to uniformly position points inside the ball.)

For this simulation method, you need to make two simple modifications to the simulation procedure.

Number of points

The number of points of a Poisson point process inside a sphere of radius $r>0$ is a Poisson random variable with mean $\lambda V_3$, where $V_3=4\pi r^3$ is the volume of the sphere.

Locations of points

We will modify Method 1 as outlined above. To sample the points uniformly in the sphere, you need to generate uniform variables on the unit interval $(0,1)$, take their cubic roots, and then, multiply them by the radius $r$. (This is akin to the step of taking the square root in the disk setting.) The random variables for the angular coordinates are generated as before.

Simulating Matérn hard-core point processes

If you wanted to create a point process with repulsion, a reasonable first attempt would be to build off a Poisson point process by removing points according to some rule to ensure that no two points were within a certain distance of each other. Using this natural idea, Bertril Matérn proposed a family of repulsive point processes called Matérn hard-core point processes.

More specifically, Matérn proposed several points processes, including two types of hard-core point processes now called Type I and Type II. (Matérn proposed a third type, called Type III, but it’s considerably harder to simulate on a computer, as detailed in this article.) These types of hard-core point processes are completely different to the Matérn cluster point process.

As I discussed in a previous post, the Poisson point process may not be adequate for representing point phenomena whose points exhibit large degrees of repulsion or clustering. I already covered the Matérn and Thomas cluster point processes, which show distinct clustering in their configurations. In this post, I’ll cover Matérn hard-core point processes. The Type I point processes is the easier of the two, so I’ll start with that one.

Overview

Simulating Matérn hard-core point processes requires first simulating a homogeneous Poisson point process with an intensity $\lambda>0$ on some simulation window, such as a rectangle, which is the simulation window I will use here. I have already written about simulating the homogeneous Poisson point processes on a rectangle and a disk, so those posts are good starting points.

Given the Poisson point process, the points then need to be thinned in such a manner to ensure that for each point, there is no other point within some fixed $r>0$ of the point. This distance $r>0$ is the radius of the hard core of each point.

I have already covered the point process operation of thinning. But it’s important to note here that in this construction a dependent thinning is being applied. (If I just applied an independent thinning, then the resulting point process will be another Poisson point process with no repulsion between points.)

Edge effects

The main trick behind sampling this point process is that it’s possible for points inside the simulation window to be thinned due to their closeness to points that are located outside the simulation window. In other words, points outside the simulation window can cause points inside the window to be thinned. (I discussed a very similar issue in the posts on the Matérn and Thomas cluster point processes.)

To remove these edge effects, the underlying Poisson point process must be simulated on an extended version of the simulation window. The points are then thinned according to a dependent thinning, which is covered in the next section. Then only the retained points inside the simulation window are kept and the remaining points are ignored. Consequently, the underling Poisson points are simulated on an extended window, but we only see the final points inside the simulation window.

To create the extended simulation window, we add a strip of width $r$ all around the simulation window. Why? Well, the distance $r$ is the maximum distance from the simulation window that another point (outside the simulation window) can exist, while still causing points inside the simulation window to be thinned. This means it is impossible for a hypothetical point beyond this distance (outside the extended window) to cause a point inside the simulation window to be thinned.

Dependent thinning rules

Type I

For each point inside the simulation window, check if there are any other points (including those in the extended window) within distance $r$ of the point. If no, then keep the point. If yes, then remove the point and the points that are within distance $r$ of the point. The remaining points inside the simulation window form a Matérn Type I point process.

This is a relatively simple thinning rule, which only requires calculating all the inter-point distances. But it is also a very strong thinning rule, meaning that it removes many points. Depending on the Poisson point process intensity $\lambda$ and core radius $r$, it is quite possible that all the points are removed, resulting in an empty configuration.

Now we examine the case when the thinning rule is not as strong.

Type II

To create Matérn Type II point process, we assign an independent uniform random variable to each point of the underlying Poisson point process defined on the extended window. In point process terminology, these random variables are called marks, resulting in a marked point process. In the the context of the Matérn Type II point process, these random random marks are usually called ages.

Then for each point in the simulation window, we consider all the points within distance $r$ of the point. If this point is the youngest (or, equivalently, the oldest) point, then the point is kept. In other words, the point is only kept if its random mark is smaller (or larger) than the random marks of all the other points within distance $r$ of the point. The remaining points inside the simulation window form a Matérn Type II point process.

Intensity expressions

Using point process and probability theory, one can derive mathematical expressions for the intensities (that is, the average density of points per unit area). These closed-form expressions can then be used to check that the correct number of points are being generated on average over many simulations.

Type I

The intensity of the Type I point process is given by

\[\mu_1=\lambda e^{-\lambda \pi r^2},\]

where $\lambda \pi r^2$ is simply the area of the core.

Type II

The intensity of the Type II point process is given by

\[\mu_2=\frac{1}{\pi r^2}(1-e^{-\lambda \pi r^2}),\]

which can be written with the intensity of the the Type I point process as

\[\mu_2=\frac{1}{\pi r^2}(1-\frac{\mu_1}{\lambda}).\]

Code

I wrote the sampling code in MATLAB and Python, which are, as usual, very similar to each other. The code, which is is located here, simulates both Type I and II Matérn points processes. It also compares the empirical intensity to the the values given by the mathematical expressions in the previous section.

MATLAB

The MATLAB code is here.

Python

The Python code is here.

Results

I have plotted single realizations of the Matern Type I and II point processes, as well as the underlying Poisson point process in the same window.

MATLAB

Python

Simulating Poisson random variables – Survey of methods

In the previous post, I discussed how to sample or generate Poisson random variables or, more correctly, variates. I detailed a direct method that uses the fact that a Poisson stochastic process, which is directly related to a Poisson point process, has inter-arrival times that form independent and identically distributed exponential variables.

This direct method in turn can be easily reformulated so it only uses (standard) uniform variables to generate Poisson random variables. It is an easy and intuitive sampling method, explaining why it is often used. Using it, I wrote Poisson simulation code in MATLAB, Python, C and C#, which can be found here. (For another post, I later implemented the same Poisson sampling method in Fortran, which is located here.)

As elegant and exact as this simulation method is, it unfortunately decreases in speed as the Poisson parameter $\lambda$ increases. In a tutorial published in 1983, Brian D. Ripely, a major figure in spatial statistics, says this about the direct method:

This is simple, but has expected time proportional to $\lambda$. Some of its competitors use rejection methods with the envelope distribution that of the integer part of a continuous random variable, such as logistic, Laplace and normal mixed with exponential distributions.

We recall that acceptance-rejection or rejections methods involve simulating a random object, such as a random variable, by first simulating another random object of the same type that is easier to simulate. The simulation method then accepts or rejects these random objects based on a certain ratio. The distribution of the simpler random object that is first simulated is called the envelope distribution. Such rejection methods are one way to simulate Poisson variables.

In short, when simulating Poisson variables, the appropriate simulation algorithm should be chosen based on the Poisson parameter. Consequently, the code of most computer functions for generating Poisson variables will have an if-statement, using the direct method for small parameter values and another method for large parameter values. We now consider the other methods.

Different methods

Over the years there have been different methods proposed for producing Poisson random variates. In the book Non-uniform random variate generation, Luc Devroye groups the different methods into five categories coupled with his views. I’ll briefly describe these methods:

Direct methods based on the homogeneous Poisson stochastic process having exponential inter-arrival times. These methods are simple, but the expected time is proportional to the Poisson parameter $\lambda$.
Inversion methods that search through a table of cumulative Poisson probabilities. Examples include the papers by Fishman (1976) and Atkinson (1979)*.
Methods that use the recursive properties of the Poisson distribution. The paper by Ahrens and Dieter (1974) uses this approach, and its expected time of completion is proportional to $\log\lambda$.
Acceptance-rejection (or rejection) methods that give relatively fast but simple algorithms. Such methods are proposed in the papers by Atkinson (1979)*, Ahrens and Dieter (1980) and Devroye (1981) or the technical report by Schmeiser and Kachitvichyanukul (1981).
Acceptance-complement methods that uses a normal distribution as the starting distribution, such as the paper by Ahrens and Dieter (1982). This method is fast, but the code is rather long.

*Atkinson had (at least) two papers on generating Poisson variates published in 1979, but I believe Devroye is referring to the first paper, because in the second paper Atkinson compares methods proposed by others.

For the paper titles, see the Further reading section below.

Methods implemented

In this section, I’ll state which proposed methods are used in various programming languages and numerical methods. I won’t go into the details how the methods work, as I’ll just cite the papers instead.

MATLAB

For small $\lambda$ values, the MATLAB function poissrnd uses the direct method (based on inter-arrival times) with a while-loop.

For $\lambda$ values greater than fifteen, I believe that the MATLAB function poissrnd uses Algorithm PG from the 1974 paper by Ahrens and Dieter. But to come to this conclusion, I had to do some investigating. You can skip to the next section if you’re not interested, but now I’ll explain my reasoning.

The MATLAB documentation says it uses a method proposed by Ahrens and Dieter, but these two researchers developed a number of methods for generating Poisson variables. The MATLAB code cites Volume 2 of the classic series by Knuth, who says the method is due to Ahrens and Dieter, but he doesn’t give an exact citation in that section of the book. Confusingly, Knuth cites in his book a couple papers by Ahrens and Dieter for generating different random variates. (Knuth later cites a seemingly relevant 1980 paper by Ahrens and Dieter, but that details another method.)

Both the MATLAB code and Knuth cite the book by Devroye. In his book (Exercise 3.5.2), Devroye discusses one method, among others, from a 1974 paper by Ahrens and Dieter. Another hint is given by examining the code of the MATLAB function poissrnd, which reveals that it uses the function randg to generate gamma variables. In the Ahrens and Dieter 1974 paper, their Algorithm PG (for producing Poisson variates) uses gamma random variables, and it’s suggested to use a parameter value of $7/8$. This is the same parameter used in the MATLAB code and mentioned by Knuth, confirming that this is the right paper by Ahrens and Dieter.

In summary, for large $\lambda$ the function MATLAB uses Algorithm PG from the 1974 paper by Ahrens and Dieter, whereas for small values it uses the direct method, which they refer to as the multiplication method.

R

In R, the function rpois use an algorithm outlined in the 1982 paper by Ahrens and Dieter. You can view the R source code here. The two cases for $\lambda$ (or $\mu$ in the paper) depend on whether $\lambda$ is greater than ten or not. For small $\lambda$, the R function rpois does not use the method based on inter-arrival times, but rather an inversion method based on a table of (cumulative) probabilities given by the Poisson probability distribution.

Python (NumPy)

In NumPy, the function numpy.random.poisson generates Poisson variates. The source code for the NumPy library is here, but for the Poisson function the underlying code is actually written in C; see the distributions.c file located here. For small Poisson parameter $\lambda$, the code uses the direct method; see the function random_poisson_mult in the code.

For Poisson parameter $\lambda \geq 10$, the comments in the code reveal that it uses a method from a 1993 paper by Hörmann; see Algorithm PTRS on page 43 of the paper. This is a transformation method, which for NumPy is implemented in the C code as the function random_poisson_ptrs. The method, which Hörmann calls the transformed rejection with squeeze, combines inversion and rejection methods.

Octave

Octave is intended to be a GNU clone of MATLAB, so you would suspect it uses the same methods as MATLAB for generating Poisson random variates. But the Octave function poissrnd uses different methods. The code reveals it generates the Poisson variates with a function called prand. It considers different cases depending on the value of the Poisson parameter $\lambda$ as well as whether a single variable (that is, a scalar) or vector or matrix of Poisson variates are being generated.

In total, the Octave function prand uses five different methods. For two of the methods, the documentation cites methods from the classic book Numerical Recipes in C (the 1992 edition); see next section. To generate a single Poisson variate with Poisson parameter $\lambda \leq 12$, the Octave function prand uses the direct method based on inter-arrival times.

Numerical Recipes (Fortran, C and C++)

The book Numerical Recipes is a classic by Press, Teukolsky, Vetterling and Flannery on numerical methods. The books comes in different editions reflecting different publication years and computer languages. (In the first two editions of the book, the authors implemented the algorithms respectively in Fortran and C.)

For generating Poisson variates, the book contents seems to have not changed over the editions that I looked at, which covered the programming languages Fortran (77 and 90), C, and C++. The authors cover Poisson generation in Section 7.3 in the Fortran and C editions. In the third edition of Numerical Recipes, they implement their methods in C++ in Section 7.3.12.

For small values of Poisson parameter $\lambda$, Numerical Recipes uses the direct method. For $\lambda >12$ values, an acceptance-rejection method is used, which relies upon finding a continuous version of the discrete Poisson probability distribution.

GSL Library (C)

In the GSL library, one can use the function gsl_ran_poisson, which uses the the direct method of exponential times. The code, which can be viewed here, cites simply Knuth (presumably the second volume).

NAG Library (C)

Although I didn’t see the code, it appears that the function nag_rand_poisson (g05tjc ) in the NAG library also uses the direct method, based on the material in the second volume of series by Knuth. But in a 1979 paper Atkinson says that the NAG library uses a method from the 1974 paper by Ahrens and Dieter.

MKL library (C)

In the MKL C library written by Intel, there seems to be three methods in use for generating Poisson variates.

The first function is called VSL_RNG_METHOD_POISSON_PTPE, which does the following for a Poisson distribution with parameter $\Lambda$:

If Λ ≥ 27, random numbers are generated by PTPE method. Otherwise, a combination of inverse transformation and table lookup methods is used. The PTPE method is a variation of the acceptance/rejection method that uses linear (on the fraction close to the distribution mode) and exponential (at the distribution tails) functions as majorizing functions. To avoid time-consuming acceptance/rejection checks, areas with zero probability of rejection are introduced and a squeezing technique is applied.

This function uses the so-called PTPE method, which is outlined in a 1981 technical report by Schmeiser and Kachitvichyanukul.

The second function is called VSL_RNG_METHOD_POISSON_POISNORM, which does the following :

If Λ < 1, the random numbers are generated by combination of inverse transformation and table lookup methods. Otherwise, they are produced through transformation of the normally distributed random numbers.

The third function is called VSL_RNG_METHOD_POISSONV_POISNORM, which does the following:

If Λ < 0.0625, the random numbers are generated by inverse transformation method. Otherwise, they are produced through transformation of normally distributed random numbers.

cuRAND (C)

Finally, Nvidia’s CUDA Random Number Generation library (cuRAND) has a function for generating Poisson variates. To see the C code, copies of it can be found in various GitHub repositories, such as this one. The cuRAND function curand_poisson uses the direct function for Poisson parameter values less than 64. For parameters values greater than 4000, it uses a normal approximation (rounded to the nearest integer).

For other values, the function curand_poisson uses a rejection method based on an approximation of the incomplete gamma function; see the function curand_poisson_gammainc. The book by Fishman is cited; see Section 8.16.

Boost library Random (C++)

The Boost library Random uses the PTRD algorithm proposed in the 1993 paper by Hörmann to generate Poisson variates; see Algorithm PTRD on page 42 of the paper. In the same paper appears the PTRS method, which is used by Python (NumPy) (though implemented in C), as mentioned above.

Simulating Poisson random variables – Direct method

If you were to write from scratch a program that simulates a homogeneous Poisson point process, the trickiest part would be the random number of points, which requires simulating a Poisson random variable. In previous posts, such as this one and this one, I’ve simply used the inbuilt functions for simulating (or generating) Poisson random variables (or variates).

But in this post I will write my own Poisson simulation code in MATLAB, Python, C and C#, which can be found here. My online webpage editor tends to mangle symbols like < and >, so it’s best not to copy my code straight from the website, unless you check and edit it.

(In the literature, researchers describe methods on how to generate random deviates or variates. But, in my informal way, I will often say simulate random variables or generate random variates, somewhat interchangeably.)

But how would one create such a Poisson function using just a standard uniform random variate generator?

The method being used depends on the value of the Poisson parameter, denoted here by $\lambda$, which is the mean (as well as the variance) of a random variable with a Poisson distribution. If this parameter value is small, then a direct simulation method can be used to generate Poisson random variates. In practice a small Poisson parameter is a number less than some number between 10 to 30.

For large $\lambda$ values, other methods are generally used, such as rejection or (highly accurate) approximation methods. In the book Non-uniform random variate generation, the author Luc Devroye groups the methods into five categories (Section X.3.2), which I briefly describe in the next post. The first of those categories covers the method that I mentioned above. I will cover that method in this post, presenting some Poisson sampling code in C and C#. (I will also present some code in MATLAB, but you would never use it instead of the the inbuilt function poissrnd.)

In the next post, I’ll describe other types of Poisson simulation methods, and I’ll detail which simulation methods various programming libraries use.

Direct method

An elegant and natural method for simulating Poisson variates is using a result based on the homogeneous Poisson stochastic process. The points in time when a given homogeneous Poisson stochastic process increases forms a Poisson point process on the real line. (Confusingly, the term Poisson process is often used to refer to both the stochastic process and the point process, but there is a slight difference.)

Using exponential random variables

The times when the Poisson stochastic process increases are called arrival times or occurrence times, as in classic stochastic models they represent the arrivals or occurrences of something, such as phone calls over time. The differences between consecutive times are called inter-arrival times or inter-occurrence times. The inter-arrival times of a homogeneous Poisson process form independent exponential random variables, a result known as the Interval Theorem.

Using this connection to the Poisson stochastic process, we can generate exponential variables $E_1$, $E_2, \dots $, and add them up. The smallest number of exponential variables for the resulting sum to exceeds one will give a Poisson random variable. That is, if we define $N$ to be the smallest $n$ such that
$$ \sum_{k=1}^n E_k > 1, $$
then $N$ is a random variable distributed according to a Poisson distribution.

Generating exponential variates is easily done by using the inverse method. For a uniform random variable $U$ on the unit interval $(0,1)$, the transformation $E= -\log(U)/\lambda $ gives an exponential random variable with mean $1/\lambda$.

Using uniform random variables

To reduce computations, the direct method using exponential random variables is often reformulated as products of uniform random variables. We can do this, due to logarithmic identities, and work with products of uniform variables instead of sums of exponential random variables. Then using standard uniform random variables $U_1, U_2,\dots$, we define $N$ to be the smallest $n$ such that
$$ \prod_{k=1}^n U_k > e^{-\lambda}. $$

These two different formulations of the same method are captured by Lemma 3.2 and Lemma 3.3 in Chapter 10 of Devroye’s book.

Example in MATLAB

In MATLAB, we can implement this method with the first formulation in a function with a simple while-loop:

function N=funPoissonLoop(lambda)
T=0; %initialize sum of exponential variables as zero
n=-1;%initialize counting variable as negative one

while (T < 1)
E=-(1/lambda)*log(rand(1));%generate exponential random variable
T=T+E; %update sum of exponential variables
n=n+1; %update number of exponential variables
end
N=n;
end

But, as I said before, don’t use this code instead of the inbuilt function poissrnd.

If you want to be a bit more tricky, you could achieve the same result by using recursion:

function N=funPoissonRecursive(lambda)
T=0; %initialize sum of exponential variables as zero
n=-1; %initialize counting variable as negative one

%run (recursive) exponential function step function
[~,N]=funStepExp(lambda,T,n);

function [T,N]=funStepExp(nu,S,m)
if (S < 1)
%run if sum of exponential variables is not high enough

%generate exponential random variable
E=(-log(rand(1)))/nu;
S=S+E; %update sum of exponential variables
m=m+1; %update nunber of exponential variables

%recursively call function again
[T,N]=funStepExp(nu,S,m);
else
T=S;
N=m;
end
end
end

Note how the code recursively calls the function funStepExp, which generates an exponential variable each time.

In the Code section below I describe my code in C and C#, using the second formulation.

Origins

Some people attribute the direct method, based on inter-arrival times, to (or, at least, cite) Donald Knuth, who details it in the second volume of his classic series of books, but I doubt that the great Knuth was the first to have this idea. For example, a quick search on Google Scholar found a paper by K. D. Tocher on computers and random sampling, where Tocher proposes the direct method in 1954, some years before Knuth started publishing his classic series.

The direct method for Poisson sampling relies upon the Interval theorem. The Poisson point process expert Günter Last studied the origins of this fundamental result. He presented its history in a recent book authored by him and Matthew Penrose; see Chapter 7 and its corresponding historical footnotes in Section C of the appendix. (A free version of the book can be found here. ) People connected to the result include Robert Ellis and William Feller.

Other methods

The direct method perfectly generates Poisson random variables (or I should say Poisson random variates). But it can be too slow for large values of the Poisson parameter (that, is the mean) $\lambda$. This has motivated researchers to develop other methods, which I will mention in the next post.

Code

I wrote some code that simulates Poisson random variables by employing the direct method based on exponential inter-arrival times. As always, all my the code is online, with the code from this post being located here.

I have implemented the second formulation (using just uniform variables) in the C and C# languages. In the code, I have used a while-loop to implement the method. But I could have also used a recursion method, as I did in the MATLAB example above, which I have also done in Python (with NumPy).

For an empirical test, the code also calculates the mean and variance of a collection of Poisson variables. For a large enough number of variables, the sample mean and the variance will closely agree with each other, converging to the same value.

C

Warning: My C code uses rand(), the standard pseudo-random number function in C, which is known for failing certain tests of randomness. The function is adequate for regular simulation work. But it gives poor results for large number of simulations. Replace this function with another pseudo-random number generator.

The code for generating a single Poisson variate is fairly straightforward in C. Here’s a sample of the code with just the Poisson function:

//Poisson function -- returns a single Poisson random variable
int funPoissonSingle(double lambda)
{
double exp_lambda = exp(-lambda); //constant for terminating loop
double randUni; //uniform variable
double prodUni; //product of uniform variables
int randPoisson; //Poisson variable

//initialize variables
randPoisson = -1;
prodUni = 1;
do
{
randUni = funUniformSingle(); //generate uniform variable
prodUni = prodUni * randUni; //update product
randPoisson++; //increase Poisson variable

} while (prodUni > exp_lambda);
return randPoisson;
}

For generating multiple variates, the code becomes more complicated, as one needs to use pointers, due to the memory capabilities of C. Again, the function uses the pseudo-random number generator in C.

C#

The code for generating a single Poisson variate is also straightforward in C#. Here’s the function in C#:

//Poisson function -- returns a single Poisson random variable
public int funPoissonSingle (double lambda) {
double exp_lambda = Math.Exp (-lambda); //constant for terminating loop
double randUni; //uniform variable
double prodUni; //product of uniform variables
int randPoisson; //Poisson variable

//initialize variables
randPoisson = -1;
prodUni = 1;
do {
randUni = funUniformSingle (); //generate uniform variable
prodUni = prodUni * randUni; //update product
randPoisson++; // increase Poisson variable

} while (prodUni > exp_lambda);

return randPoisson;
}

Generalizing the code so it generates multiple variates just requires a little change, compared to C, as the C# language is a much more modern language.

Simulating Poisson point processes faster

As an experiment, I tried to write code for simulating many realizations of a homogeneous Poisson point process in a very fast fashion. My idea was to simulate all the realizations in two short steps.

In reality, the findings of this experiment and the contents of this post have little practical value, as computers are so fast at generating Poisson point processes. Still, it was an interesting task, which taught me a couple of things. And I did produce faster code.

MATLAB

I first tried this experiment in MATLAB.

Vectorization

In languages like MATLAB, the trick for speed is to use vectorization, which means applying a single operation to an entire vector or matrix (or array) without using an explicit for-loop. Over the years, the people behind MATLAB has advised to use vectorization instead of for-loops, as for-loops considerably slowed down MATLAB code. (But, as as time goes by, it seems using for-loops in MATLAB doesn’t slow the code down as much as it used to.)

Simulating Poisson point processes is particularly amenable to vectorization, due to the independent nature of the point process. One can simulate the number of points in each realization for all realizations in one step. Then all the points across all realizations can also be positioned in one step. In the two-dimensional case, this results in two one-dimensional arrays (or vectors, in MATLAB parlance) for the $x$ and $y$ coordinates. (Of course, in my code, I could have used just one two-dimensional array/vector for the coordinates of the points, but I didn’t.)

After generating the points, the coordinates of the points need to be grouped into the different realizations and stored in appropriate data structures.

Data structures

The problem with storing point processes is that usually each realization has a different number of points, so more sophisticated data structures than regular arrays are needed. For MATLAB, each point process realization can be stored in a data object called a cell array. For a structure array, it’s possible for each element (that is, structure) to be a different size, making them ideal for storing objects like point processes with randomly varying sizes.

In the case of two-dimensional point processes, two cell arrays can be used to store the $x$ and $y$ coordinates of all the point process realizations. After randomly positioning all the points, they can be grouped into a cell array, where each cell array element represents a realization of the Poisson point process, by using the inbuilt function MATLAB mat2cell, which converts a matrix (or array) into a cell array.

Alternatively, we could use another MATLAB data object called a structure array. In MATLAB structures have fields, which can be, for example for a point process can be the locations of the points. Given cell arrays of equal size, we can convert them into a single structure array by using the inbuilt MATLAB function struct.

Python

After successfully simulating Poisson point processes in MATLAB, I tried it in Python with the NumPy library.

Vectorization

I basically replicated what I did in MATLB using Python by positioning all the points in a single step. This gives two one-dimensional NumPy arrays for the $x$ and $y$ coordinates of all the point process realizations. (Again, I could have instead stored the coordinates as a single two-dimensional array, but I didn’t.)

Perhaps surprisingly, the vectorization step doesn’t speed things up much in Python with the NumPy library. This may be due to the fact that the underlying code is actually written in the C language. That motivated me to see what methods have been implemented for simulating Poisson variables, which is the topic of the next couple posts.

Data structures

In Python, the data structure known as a list is the natural choice. Similarly to cell arrays in MATLAB, two lists can be used for the $x$ and $y$ coordinates of all the points. Instead of MATLAB’s function mat2cell, I used the NumPy function numpy.split to create two lists from the two NumPy arrays containing the point coordinates.

Python does not seem to have an immediate equivalent to structure arrays in MATLAB. But in Python one can define a new data structure or class with fields, like a structure. Then one can create a list of those data structures with fields, which are called attribute references in Python.

Code

The code in MATLAB and Python can be found here. For a comparison, I also generated the same number of point process realizations (without using vectorization) by using a trusty for-loop. The code compares the times of taken for implemented the two different approaches, labelled internally as Method A and Method B. There is a some time difference in the MATLAB code, but not much of a difference in the Python case.

I have commented out the sections that create data structures (with fields or attribute references) for storing all the point process realizations, but those sections should also work when uncommented.

Simulating a Cox point process based on a Poisson line process

In the previous post, I described how to simulate a Poisson line process, which in turn was done by using insight from an earlier post on the Bertrand paradox.

Now, given a Poisson line process, for each line, if we generate an independent one-dimensional Poisson point point process on each line, then we obtain an example of a Cox point process. Cox point processes are also known as doubly stochastic Poisson point processes. On the topic of names, Guttorp and Thorarinsdottir argue that it should be called the Quenouille point process, as Maurice Quenouille introduced an example of it before Sir David Cox, but I opt for the more common name.

Cox point proceesses

A Cox point process is a generalization of a Poisson point process. It is created by first considering a non-negative random measure, sometimes called a driving measure. Then a Poisson point process, which is independent of the random driving measure, is generated by using the random measure as its intensity or mean measure.

The driving measure of a Cox point process can be, for example, a non-negative random variable or field multiplied by a Lebesgue measure. In our case, the random measure is the underlying Poisson line process coupled with the Lebesgue measure on the line (that is, length).

Cox processes form a very large and general family of point processes, which exhibit clustering. In previous posts, I have covered two special cases of Cox point processes: the Matérn and Thomas cluster point processes. These are, more specifically, examples of a Neyman-Scott point process, which is a special case of a shot noise Cox point process. These two point processes are fairly easy to simulate, but that’s not the case for Cox point processes in general. Some are considerably easier than others.

Motivation

I will focus on simulating the Cox point process formed from a Poisson line process with homogeneous Poisson point processes. I do this for two reasons. First, it’s easy to simulate, given we can simulate a Poisson line process. Second, it has been used and studied recently in the mathematics and engineering literature for investigating wireless communication networks in cities, where the streets correspond to Poisson lines; for example, see these two preprints:

Incidentally, I don’t know what to call this particular Cox point process. A Cox line-point process? A Cox-Poisson line-point process? But it doesn’t matter for simulation purposes.

Method

We will simulate the Cox (-Poisson line-) point process on a disk. Why a disk? I suggest reading the previous posts on the Poisson line process the Bertrand paradox for why the disk is a natural simulation window for line processes.

Provided we can simulate a Poisson line process, the simulation method is quite straightforward, as I have essentially already described it.

Line process

First simulate a Poisson line process on a disk. We recall that for each line of the line process, we need to generate two independent random variables $\Theta$ and $P$ describing the position of the line. The first random variable $\Theta$ gives the line orientation, and it is a uniform random variable on the interval $(0,2\pi)$.

The second random variables $P$ gives the distance from the origin to the disk edge, and it is a uniform random variable on the interval $(0,r)$, where $r$ is the radius of the disk. The distance from the point $(\Theta, P)$ to the disk edge (that is, the circle) along the chord is:

$$Q=\sqrt{r^2-P^2}.$$

The endpoints of the chord (that is, the points on the disk edge) are then:

Point 1: $X_1=P \cos \Theta+ Q\sin \Theta$, $Y_1= P \sin \Theta- Q\cos \Theta$,

Point 2: $X_2=P \cos \Theta- Q\sin \Theta$, $Y_2= P \sin \Theta+Q \cos \Theta$.

The length of the line segment is $2 Q$. We can say this random line is described by the point $(\Theta,P)$.

One-dimensional Poisson point process

For each line (segment) in the line process, simulate a one-dimensional Poisson point process on it. Although I have never discussed how to simulate a one-dimensional (homogeneous) Poisson point process, it’s essentially one dimension less than simulating a homogeneous Poisson point process on a rectangle.

More specifically, given a line segment $(\Theta,P)=(\theta,p)$, you simulate a homogeneous Poisson point process with intensity $\mu$ on a line segment with length $2 q$, where $q=\sqrt{r^2-p^2}$. (I am now using lowercase letters to stress that the line is no longer random.) To simulate the homogeneous Poisson point process, you generate a Poisson random variable with parameter $2 \mu q$.

Now you need to place the points uniformly on the line segment. To do this, consider a single point on a single line. For this point, generate a single uniform variable $U$ on the interval $(-1,1)$. The tricky part is now getting the Cartesian coordinates right. But the above expressions for the endpoints suggest that the single random point has the Cartesian coordinates:

$x=p \cos \theta+ U q\sin \theta$, $y=p \sin \theta- U q\cos \theta$.

The two extreme cases of the uniform random variable $U$ (that is, $U=-1$ and $U=1$) correspond to the two endpoints of the line segment. We recall that $Q$ is the distance from the midpoint of the line segment to the disk edge along the line segment, so it makes sense that we want to vary this distance uniformly in order to uniformly place a point on the line segment. This uniform placement step is done for all the points of the homogeneous Point process on that line segment.

You repeat this procedure for every line segment. And that’s it: a Cox point process built upon a Poisson line process.

Results

MATLAB

R

Python

Code

As always, the code from all my posts is online. For this post, I have written the code in MATLAB, R and Python.

Simulating a Poisson line process

Instead of points, we can consider other objects randomly scattered on some underlying mathematical space. If we take a Poisson point process, then we can use (infinitely long) straight lines instead of points, giving a Poisson line process. Researchers have studied and used this random object to model physical phenomena. In this post I’ll cover how to simulate a homogeneous Poisson line process in MATLAB, R and Python. The code can be downloaded from here.

Overview

For simulating a Poisson line process, the key question is how to randomly position the lines. This is related to a classic problem in probability theory called the Bertrand paradox. I discussed this illustration in probability in a previous post, where I also included code for simulating it. I highly recommend reading that post first.

The Bertrand paradox involves three methods for positioning random lines. We use Method 2 to achieve a uniform positioning of lines, meaning the number of lines and orientation of lines is statistically uniform. Then it also makes sense to use this method for simulating a homogeneous (or uniform) Poisson line process.

We can interpret a line process as a point process. For a line process on the plane $\textbf{R}^2$, it can be described by a point process on $(0,\infty)\times (0,2\pi)$, which is an an infinitely long cylinder. In other words, the Poisson line process can be described as a Poisson point process.

For simulating a Poisson line process, it turns out the disk is the most natural setting. (Again, this goes back to the Bertrand paradox.) More specifically, how the (infinitely long) lines intersect a disk of a fixed radius $r>0$. The problem of simulating a Poisson line process reduces to randomly placing chords in a disk. For other simulation windows in the plane, we can always bound any non-circular region with a sufficiently large disk.

Steps

Number of lines

Of course, with all things Poisson, the number of lines will be a Poisson random variable, but what will its parameter be? This homogeneous (or uniform) Poisson line process forms a one-dimensional homogeneous (or uniform) Poisson point process around the edge of the disk with a circumference $2 \pi r $. Then the number of lines is simply a Poisson variable with parameter $\lambda 2 \pi r $.

Locations of points

To position a single line uniformly in a disk, we need to generate two uniform random variables. One random variable gives the angle describing orientation of the line, so it’s a uniform random variable $\Theta$ on the interval $(0,2\pi)$.

The other random variable gives the distance from the origin to the disk edge, meaning it’s a uniform random variable $P$ on the interval $(0,r)$, where $r$ is the radius of the disk. The random radius and its perpendicular chord create a right-angle triangle. The distance from the point $(\Theta, P)$ to the disk edge (that is, the circle) along the chord is:

$$Q=\sqrt{r^2-P^2}.$$

The endpoints of the chord (that is, the points on the disk edge) are then:

Point 1: $X_1=P \cos \Theta+ Q\sin \Theta$, $Y_1= P \sin \Theta- Q\cos \Theta$,

Point 2: $X_2=P \cos \Theta- Q\sin \Theta$, $Y_2= P \sin \Theta+Q \cos \Theta$.

Code

I have implemented the simulation procedure in MATLAB, R and Python, which, as usual, are all very similar. I haven’t put my code here, because the software behind my website keeps mangling it. As always, I have uploaded my code to a repository; for this post, it’s in this directory.

I have written the code in R, but I wouldn’t use it in general. That’s because if you’re using R, then, as I have said before, I strongly recommend using the powerful spatial statistics library spatstat. For a simulating Poisson line process, there’s the function rpoisline.

The chief author of spatstat, Adrian Baddeley, has written various lectures and books on the related topics of point processes, spatial statistics, and geometric probability. In this post, he answered why the angular coordinates have to be chosen uniformly.

Results

MATLAB

Python

The Bertrand paradox

Mathematical paradoxes are results or observations in mathematics that are (seemingly) conflicting, unintuitive, incomprehensible, or just plain bizarre. They come in different flavours, such as those that play with notions of infinity, which means they often make little or no sense in a physical world. Other paradoxes, particularly those in probability, serve as a lesson that the problem needs to be posed in a precise manner. The Bertrand paradox is one of these.

Joseph Bertrand posed the original problem in his 1889 book Calcul des probabilités, which is available online (albeit in French); page 4, Section 5. It’s a great illustrative problem involving simple probability and geometry, so it often appears in literature on the (closely related) mathematical fields of geometric probability and integral geometry.

Based on constructing a random chord in a circle, Bertrand’s paradox involves a single mathematical problem with three reasonable but different solutions. It’s less a paradox and more a cautionary tale. It’s really asking the question: What exactly do you mean by random?

Consequently, over the years the Bertrand paradox has inspired debate, with papers arguing what the true solution is. I recently discovered it has even inspired some passionate remarks on the internet; read the comments at www.bertrands-paradox.com.

Update: The people from Numberphile and 3Blue1Brown have recently produced a video on YouTube describing and explaining the Bertrand paradox.

But I am less interested in the different interpretations or philosophies of the problem. Rather, I want to simulate the three solutions. This is not very difficult, provided some trigonometry and knowledge from a previous post, where I describe how to simulate a (homogeneous) Poisson point process on the disk.

I won’t try to give a thorough analysis of the solutions, as there are much better websites doing that. For example, this MIT website gives a colourful explanation with pizza and fire-breathing monsters. The Wikipedia article also gives a detailed though less creative explanation for the three solutions.

My final code in MATLAB, R and Python code is located here.

The Problem

Bertrand considered a circle with an equilateral triangle inscribed it. If a chord in the circle is randomly chosen, what is the probability that the chord is longer than a side of the equilateral triangle?

The Solution(s)

Bertrand argued that there are three natural but different methods to randomly choose a chord, giving three distinct answers. (Of course, there are other methods, but these are arguably not the natural ones that first come to mind.)

Method 1: Random endpoints

On the circumference of the circle two points are randomly (that is, uniformly and independently) chosen, which are then used as the two endpoints of the chord.

The probability of this random chord being longer than a side of the triangle is one third.

Method 2: Random radius

A radius of the circle is randomly chosen (so the angle is chosen uniformly), then a point is randomly (also uniformly) chosen along the radius, and then a chord is constructed at this point so it is perpendicular to the radius.

The probability of this random chord being longer than a side of the triangle is one half.

Method 3: Random midpoint

A point is randomly (so uniformly) chosen in the circle, which is used as the midpoint of the chord, and the chord is randomly (also uniformly) rotated.

The probability of this random chord being longer than a side of the triangle is one quarter.

Simulation

All three answers involve randomly and independently sampling two random variables, and then doing some simple trigonometry. The setting naturally inspires the use of polar coordinates. I assume the circle has a radius $r$ and a centre at the origin $o$. I’ll arbitrarily number the end points one and two.

In all three solutions we need to generate uniform random variables on the interval $(0, 2\pi)$ to simulate random angles. I have already done this a couple of times in previous posts such as this one.

Method 1: Random endpoints

This is probably the most straightforward solution to simulate. We just need to simulate two uniform random variables $\Theta_1$ and $\Theta_2$ on the interval $(0, 2\pi)$ to describe the angles of the two points.

The end points of the chord (in Cartesian coordinates) are then simply:

Point 1: $X_1=r \cos \Theta_1$, $Y_1=r \sin \Theta_1$,

Point 2: $X_2=r \cos \Theta_2$, $Y_2=r \sin \Theta_2$.

Method 2: Random radius

This method also involves generating two uniform random variables. One random variable $\Theta$ is for the angle, while the other $P$ is the random radius, which means generating the random variable $P$ on the interval $(0, r)$.

I won’t go into the trigonometry, but the random radius and its perpendicular chord create a right-angle triangle. The distance from the point $(\Theta, P)$ to the circle along the chord is:

$$Q=\sqrt{r^2-P^2}.$$

The endpoints of the chord are then:

Point 1: $X_1=P \cos \Theta+ Q\sin \Theta$, $Y_1= P \sin \Theta- Q\cos \Theta$,

Point 2: $X_2=P \cos \Theta- Q\sin \Theta$, $Y_2= P \sin \Theta+Q \cos \Theta$.

Take note of the signs in these expressions.

Method 3: Random midpoint

This method requires placing a point uniformly on a disk, which is also done when simulating a homogeneous Poisson point process on a disk, and requires two random variables $\Theta’$ and $P’$. Again, the angular random variable $\Theta’$ is uniform.

The other random variable $P’$ is not uniform. For $P’$, we generate a random uniform variable on the unit interval $(0,1)$, and then we take the square root of it. We then multiply it by the radius, generating a random variable between $0$ and $r$. (We must take the square root because the area element of a sector is proportional to the radius squared, and not the radius.) The distribution of this random variable is an example of the triangular distribution.

The same trigonometry from Method 2 applies here, which gives the endpoints of the chord as:

Point 1: $X_1=P’ \cos \Theta’+ Q’\sin \Theta’$, $Y_1= P’ \sin \Theta’- Q’\cos \Theta’$,

Point 2: $X_2=P’\cos \Theta’- Q’\sin \Theta’$, $Y_2= P’\sin \Theta’+Q’ \cos \Theta’$,

where $Q’:=\sqrt{r^2-{P’}^2}$. Again, take note of the signs in these expressions.

Results

To illustrate how the three solutions are different, I’ve plotted a hundred random line segments and their midpoints side by side. Similar plots are in the Wikipedia article.

Method 1: Random endpoints

Method 2: Random radius

Method 3: Random midpoint

Conclusion

For the chord midpoints, we know and can see that Method 3 gives uniform points, while Method 2 has a concentration of midpoints around the circle centre. Method 1 gives results that seem to somewhere between Method 2 and 3 in terms of clustering around the circle centre.

For the chords, we see that Method 3 results in fewer chords passing through the circle centre. Methods 1 and 2 seem to give a similar number of lines passing through this central region.

It’s perhaps hard to see, but it can be shown that Method 2 gives the most uniform results. By this, I mean that the number of lines and their orientations statistically do not vary in different regions of the circle.

We can now position random lines in uniform manner. All we need now is a Poisson number of lines to generate something known as a Poisson line process, which will be the subject of the next post.

Code

The MATLAB, R and Python code can be found here. In the code, I have labelled the methods A, B and C instead 1, 2 and 3.

Simulating an inhomogeneous Poisson point process

In previous posts I described how to simulate homogeneous Poisson point processes on a rectangle, disk and triangle. But here I will simulate an inhomogeneous or nonhomogeneous Poisson point process. (Both of these terms are used, where the latter is probably more popular, but I prefer the former.) For such a point process, the points are not uniformly located on the underlying mathematical space on which the Poisson process is defined. This means that certain regions will, on average, tend to have more (or less) points than other regions of the underlying space.

Basics

Any Poisson point process is defined with a non-negative measure called the intensity or mean measure. I make the standard assumption that the intensity measure $\Lambda$ has a derivative $\lambda(x,y)$. (I usually write a single $x$ to denote a point on the plane, that is $x\in \mathbb{R}^2$, but in this post I will write the $x$ and $y$ and coordinates separately.) The function $\lambda(x,y)$ is often called the intensity function or just intensity, which I further assume is bounded, so $\lambda(x,y)<\infty$ for all points in a simulation window $W$. Finally, I assume that the simulation window $W$ is a rectangle, but later I describe how to lift that assumption.

Number of points

To simulate a point process, the number of points and the point locations in the simulation window $W$ are needed. For any Poisson point process, the number of points is a Poisson random variable with a parameter (that is, a mean) $\Lambda(W)$, which under our previous assumptions is given by the integral

$$\Lambda(W)=\int_W \lambda(x,y)dxdy. $$

Assuming we can evaluate such an integral analytically or numerically, then the number of points is clearly not difficult to simulate.

Locations of points

The difficulty lies in randomly positioning the points. But a defining property of the Poisson point process is its independence, which allows us to treat each point completely independently. Positioning each point then comes down to suitably simulating two (or more) random variables for Poisson point processes in two (or higher) dimensions. Similarly, the standard methods used for simulating continuous random variables can be applied to simulating random point locations of a Poisson point process.

In theory, you can rescale the intensity function with the total measure of the simulation window, giving

$$f(x,y):=\frac{\lambda(x,y)}{\Lambda(W)}. $$

We can then interpret this rescaled intensity function $f(x,y)$ as the joint probability density of two random variables $X$ and $Y$, because it integrates to one,

$$\int_W f(x,y)dxdy=1.$$

Clearly the method for simulating an inhomogeneous Poisson point process depends on the nature of the intensity function. For the inhomogeneous case, the random variables $X$ and $Y$ are, in general, not independent.

Transformation

To simulate an inhomogeneous Poisson point process, one method is to first simulate a homogeneous one, and then suitably transform the points according to deterministic function. For simple random variables, this transformation method is quick and easy to implement, if we can invert the probability distribution. For example, a uniform random variable $U$ defined on the interval $(0,1)$ can be used to give an exponential random variable by applying the transformation $h(u)= -(1/\lambda)\log(u)$, where $\lambda>0$, meaning $h(U)$ is an exponential random variable with parameter $\lambda>0$ (or mean $1/\lambda$).

Similarly for Poisson point processes, the transformation approach is fairly straightforward in a one-dimensional setting, but generally doesn’t work easily in two (or higher) dimensions. The reason being that often we cannot simulate the random variables $X$ and $Y$ independently, which means, in practice, we need first to simulate one random variable, then the other.

It is a bit easier if we can re-write the rescaled intensity function or joint probability density $f(x,y)$ as a product of single-variable functions $f_X(x)$ and $f_Y(y)$, meaning the random variables $X$ and $Y$ are independent. We can then simulate independently the random variables $X$ and $Y$, corresponding to the $x$ and $y$ coordinates of the points. But this would still require integrating and inverting the functions.

Markov chain Monte Carlo

A now standard way to simulate jointly distributed random variables is to use Markov chain Monte Carlo (MCMC), which we can also use to simulate the the $X$ and $Y$ random variables. Applying MCMC methods is simply applying random point process operations repeatedly to all the points. But this is a bit too tricky and involved. Instead I’ll use a general yet simpler method based on thinning.

Thinning

The thinning method is the arguably the simplest and most general way to simulate an inhomogeneous Poisson point process. If you’re unfamiliar with thinning, I recommend my previous post on thinning and the material I cite.

This simulation method is simply a type of acceptance-rejection method for simulating random variables. More specifically, it is the acceptance-rejection or rejection method, attributed to the great John von Neumann, for simulating a continuous random variable, say $X$, with some known probability density $f(x)$. The method accepts/retains or rejects/thins the outcome of each random variable/point depending on the outcome of a uniform random variable associated with each random variable/point.

The thinning or acceptance-rejection method is also appealing because it is an example of a perfect simulation method, which means the distribution of the simulated random variables or points will not be an approximation. This can be contrasted with typical MCMC methods, which, in theory, reach the desired distribution of the random variables in infinite time, which is clearly not possible in practice.

Simulating the homogeneous Poisson point process

First simulate a homogeneous Poisson point process with intensity value $\lambda^*$, which is an upper bound of the intensity function $\lambda(x,y)$. The simulation step is the easy part, but what value is $\lambda^*$?

I will use the maximum value that the intensity function $\lambda(x,y)$ takes, which I denote by

$$ \lambda_{\max}:=\max_{(x,y)\in W}\lambda(x,y),$$

so I set $\lambda^*=\lambda_{\max}$. Of course with $\lambda^*$ being an upper bound, you can use any larger $\lambda$-value, so $\lambda^*\geq \lambda_{\max}$, but that just means more points will need to be thinned.

Scientific programming languages have implemented algorithms that find or estimate minima of mathematical functions, meaning such an algorithm just needs to find the $(x,y)$ point that gives the minimum value of $-\lambda(x,y)$, which corresponds to the maximum value of $\lambda(x,y)$. What is very important is that the minimization procedure can handle constraints on the $x$ and $y$ values, which in our case of a rectangular simulation window $W$ are sometimes called box constraints.

Thinning the Poisson point process

All we need to do now is to thin the homogeneous Poisson point process with the thinning probability function

$$1-p(x,y)=\frac{\lambda(x,y)}{\lambda^*}.$$

This will randomly remove the points so the remaining points will form a inhomogeneous Poisson point process with intensity function
$$ (1-p(x,y))\lambda^* =\lambda(x).$$
As a result, we can see that provided $\lambda^*\geq \lambda_{\max}>0$, this procedure will give the right intensity function $\lambda(x,y)$. I’ll skip the details on the point process still being Poisson after thinning, as I have already covered this in the thinning post.

Empirical check

You can run an empirical check by simulating the point process a large number (say $10^3$ or $10^4$) of times, and collect statistics on the number of points. As the number of simulations increases, the average number of points should converge to the intensity measure $\Lambda(W)$, which is given by (perhaps numerically) evaluating the integral

$$\Lambda(W)=\int_W \lambda(x,y)dxdy.$$

This is a test for the intensity measure, a type of first moment, which will work for the intensity measure of any point process. But for Poisson point processes, the variance of the number of points will also converge to intensity measure $\Lambda(W)$, giving a second empirical test based on second moments.

An even more thorough test would be estimating an empirical distribution (that is, performing and normalizing a histogram) on the number of points. These checks will validate the number of points, but not the positioning of the points. In my next post I’ll cover how to perform these tests.

Results

The homogeneous Poisson point process with intensity function $\lambda(x)=100\exp(-(x^2+y^2)/s^2)$, where $s=0.5$. The results look similar to those in the thinning post, where the thinned points (that is, red circles) are generated from the same Poisson point process as the one that I have presented here.

MATLAB

Python

Method extensions

We can extend the thinning method for simulating inhomogeneous Poisson point processes a couple different ways.

Using an inhomogeneous Poisson point process

The thinning method does not need to be applied to a homogeneous Poisson point process with intensity $\lambda^*$. In theory, we could have simulated a suitably inhomogeneous Poisson point process with intensity function $\lambda^*(x,y)$, which has the condition

$$ \lambda^*(x,y)\geq \lambda(x,y), \quad \forall (x,y)\in W .$$

Then this Poisson point process is thinned. But then we would still need to simulate the underlying Poisson point process, which often would be as difficult to simulate.

Partitioning the simulation window

Perhaps the intensity of the Poisson point process only takes two values, $\lambda_1$ and $\lambda_2$, and the simulation window $W$ can be nicely divided or partitioned into two disjoints sets $B_1$ and $B_2$ (that is, $B_1\cap B_2=\emptyset$ and $B_1\cup B_2=W$), corresponding to the subregions of the two different intensity values. The Poisson independence property allows us to simulate two independent Poisson point processes on the two subregions.

This approach only works for a piecewise constant intensity function. But if if the intensity function $\lambda(x)$ varies wildly, the simulation window can be partitioned into subregions $B_1\dots,B_m$ for different ranges of the intensity function $\lambda(x)$. This allows us to simulate independent homogeneous Poisson point processes with different densities $\lambda^*_1\dots, \lambda^*_m$, where for each subregion $B_i$ we set

$$ \lambda^*_i:=\max_{x\in B_i}\lambda(x,y).$$

The resulting Poisson point processes are then suitably thinned, resulting in a more efficient simulation method. (Although I imagine the gain would often be small.)

Non-rectangular simulation windows

If you want to simulate on non-rectangular regions, which is not a disk or triangle, then the easiest way is to simulate a Poisson point process on a rectangle $R$ that completely covers the simulation window, so $W \subset R\subset \mathbb{R}^2$, and then set the intensity function $\lambda $ to zero for the region outside the simulation window $W$, that is $\lambda (x,y)=0$ when $(x,y)\in R\setminus W$.

Code

All code from my posts, as always, can be found on the my GitHub repository. The code for this post is located here. You can see that the code is very similar to that of the thinning code, which served as the foundation for this code. (Note how we now keep the points, so in the code the > has become < on the line where the uniform variables are generated).

I have implemented the code in MATLAB and Python with an intensity function $\lambda(x,y)=100\exp(-(x^2+y^2)/s^2)$, where $s>0$ is a scale parameter. Note that in the minimization step, the box constraints are expressed differently in MATLAB and Python: MATLAB first takes the minimum values then the maximum values, whereas Python first takes the $x$-values then the $y$-values.

The code presented here does not have the empirical check, which I described above, but it is implemented in the code located here. For the parameters used in the code, the total measure is $\Lambda(W)\approx 77.8068$, meaning each simulation will generate on average almost seventy-eight points.

I have stopped writing code in R for a couple of reasons, but mostly because anything I could think of simulating in R can already be done in the spatial statistics library spatstat. I recommend the book Spatial Point Patterns, co-authored by the spatstat’s main contributor, Adrian Baddeley.

MATLAB

I have used the fmincon function to find the point that gives the minimum of $-\lambda(x,y)$.

%Simulation window parameters
xMin=-1;xMax=1;
yMin=-1;yMax=1;
xDelta=xMax-xMin;yDelta=yMax-yMin; %rectangle dimensions
areaTotal=xDelta*yDelta; %area of rectangle

s=0.5; %scale parameter

%Point process parameters
fun_lambda=@(x,y)(100*exp(-((x).^2+(y).^2)/s^2));%intensity function

%%%START -- find maximum lambda -- START %%%
%For an intensity function lambda, given by function fun_lambda,
%finds the maximum of lambda in a rectangular region given by
%[xMin,xMax,yMin,yMax].
funNeg=@(x)(-fun_lambda(x(1),x(2))); %negative of lambda
%initial value(ie centre)
xy0=[(xMin+xMax)/2,(yMin+yMax)/2];%initial value(ie centre)
%Set up optimization step
options=optimoptions('fmincon','Display','off');
%Find largest lambda value
[~,lambdaNegMin]=fmincon(funNeg,xy0,[],[],[],[],...
[xMin,yMin],[xMax,yMax],'',options);
lambdaMax=-lambdaNegMin;
%%%END -- find maximum lambda -- END%%%

%define thinning probability function
fun_p=@(x,y)(fun_lambda(x,y)/lambdaMax);

%Simulate Poisson point process
numbPoints=poissrnd(areaTotal*lambdaMax);%Poisson number of points
xx=xDelta*(rand(numbPoints,1))+xMin;%x coordinates of Poisson points
yy=xDelta*(rand(numbPoints,1))+yMin;%y coordinates of Poisson points

%calculate spatially-dependent thinning probabilities
p=fun_p(xx,yy);

%Generate Bernoulli variables (ie coin flips) for thinning
booleRetained=rand(numbPoints,1)<p; %points to be thinned

%x/y locations of retained points
xxRetained=xx(booleRetained); yyRetained=yy(booleRetained);

%Plotting
plot(xxRetained,yyRetained,'bo'); %plot retained points
xlabel('x');ylabel('y');

The box constraints for the optimization step were expressed as:

[xMin,yMin],[xMax,yMax]

Python

I have used the minimize function in SciPy.

import numpy as np; #NumPy package for arrays, random number generation, etc
import matplotlib.pyplot as plt #For plotting
from scipy.optimize import minimize #For optimizing
from scipy import integrate

#Simulation window parameters
xMin=-1;xMax=1;
yMin=-1;yMax=1;
xDelta=xMax-xMin;yDelta=yMax-yMin; #rectangle dimensions
areaTotal=xDelta*yDelta;

s=0.5; #scale parameter

#Point process parameters
def fun_lambda(x,y):
   return 100*np.exp(-(x**2+y**2)/s**2); #intensity function

###START -- find maximum lambda -- START ###
#For an intensity function lambda, given by function fun_lambda,
#finds the maximum of lambda in a rectangular region given by
#[xMin,xMax,yMin,yMax].
def fun_Neg(x):
   return -fun_lambda(x[0],x[1]); #negative of lambda

xy0=[(xMin+xMax)/2,(yMin+yMax)/2];#initial value(ie centre)
#Find largest lambda value
resultsOpt=minimize(fun_Neg,xy0,bounds=((xMin, xMax), (yMin, yMax)));
lambdaNegMin=resultsOpt.fun; #retrieve minimum value found by minimize
lambdaMax=-lambdaNegMin;
###END -- find maximum lambda -- END ###

#define thinning probability function
def fun_p(x,y):
   return fun_lambda(x,y)/lambdaMax;

#Simulate a Poisson point process
numbPoints = np.random.poisson(lambdaMax*areaTotal);#Poisson number of points
xx = np.random.uniform(0,xDelta,((numbPoints,1)))+xMin;#x coordinates of Poisson points
yy = np.random.uniform(0,yDelta,((numbPoints,1)))+yMin;#y coordinates of Poisson points

#calculate spatially-dependent thinning probabilities
p=fun_p(xx,yy);

#Generate Bernoulli variables (ie coin flips) for thinning
booleRetained=np.random.uniform(0,1,((numbPoints,1)))<p; #points to be thinned

#x/y locations of retained points
xxRetained=xx[booleRetained]; yyRetained=yy[booleRetained];

#Plotting
plt.scatter(xxRetained,yyRetained, edgecolor='b', facecolor='none', alpha=0.5 );
plt.xlabel("x"); plt.ylabel("y");
plt.show();

The box constraints were expressed as:

(xMin, xMax), (yMin, yMax)

Julia

After writing this post, I later wrote the code in Julia. The code is here and my thoughts about Julia are here.