g05 Chapter Contents
g05 Chapter Introduction
NAG Library Manual

# NAG Library Function Documentnag_rand_field_2d_user_setup (g05zqc)

## 1  Purpose

nag_rand_field_2d_user_setup (g05zqc) performs the setup required in order to simulate stationary Gaussian random fields in two dimensions, for a user-defined variogram, using the circulant embedding method. Specifically, the eigenvalues of the extended covariance matrix (or embedding matrix) are calculated, and their square roots output, for use by nag_rand_field_2d_generate (g05zsc), which simulates the random field.

## 2  Specification

 #include #include
void  nag_rand_field_2d_user_setup (const Integer ns[], double xmin, double xmax, double ymin, double ymax, const Integer maxm[], double var,
 void (*cov2)(double x, double y, double *gamma, Nag_Comm *comm),
Nag_Parity parity, Nag_EmbedPad pad, Nag_EmbedScale corr, double lam[], double xx[], double yy[], Integer m[], Integer *approx, double *rho, Integer *icount, double eig[], Nag_Comm *comm, NagError *fail)

## 3  Description

A two-dimensional random field $Z\left(\mathbf{x}\right)$ in ${ℝ}^{2}$ is a function which is random at every point $\mathbf{x}\in {ℝ}^{2}$, so $Z\left(\mathbf{x}\right)$ is a random variable for each $\mathbf{x}$. The random field has a mean function $\mu \left(\mathbf{x}\right)=𝔼\left[Z\left(\mathbf{x}\right)\right]$ and a symmetric positive semidefinite covariance function $C\left(\mathbf{x},\mathbf{y}\right)=𝔼\left[\left(Z\left(\mathbf{x}\right)-\mu \left(\mathbf{x}\right)\right)\left(Z\left(\mathbf{y}\right)-\mu \left(\mathbf{y}\right)\right)\right]$. $Z\left(\mathbf{x}\right)$ is a Gaussian random field if for any choice of $n\in ℕ$ and ${\mathbf{x}}_{1},\dots ,{\mathbf{x}}_{n}\in {ℝ}^{2}$, the random vector ${\left[Z\left({\mathbf{x}}_{1}\right),\dots ,Z\left({\mathbf{x}}_{n}\right)\right]}^{\mathrm{T}}$ follows a multivariate Normal distribution, which would have a mean vector $\stackrel{~}{\mathbf{\mu }}$ with entries ${\stackrel{~}{\mu }}_{i}=\mu \left({\mathbf{x}}_{i}\right)$ and a covariance matrix $\stackrel{~}{C}$ with entries ${\stackrel{~}{C}}_{ij}=C\left({\mathbf{x}}_{i},{\mathbf{x}}_{j}\right)$. A Gaussian random field $Z\left(\mathbf{x}\right)$ is stationary if $\mu \left(\mathbf{x}\right)$ is constant for all $\mathbf{x}\in {ℝ}^{2}$ and $C\left(\mathbf{x},\mathbf{y}\right)=C\left(\mathbf{x}+\mathbf{a},\mathbf{y}+\mathbf{a}\right)$ for all $\mathbf{x},\mathbf{y},\mathbf{a}\in {ℝ}^{2}$ and hence we can express the covariance function $C\left(\mathbf{x},\mathbf{y}\right)$ as a function $\gamma$ of one variable: $C\left(\mathbf{x},\mathbf{y}\right)=\gamma \left(\mathbf{x}-\mathbf{y}\right)$. $\gamma$ is known as a variogram (or more correctly, a semivariogram) and includes the multiplicative factor ${\sigma }^{2}$ representing the variance such that $\gamma \left(0\right)={\sigma }^{2}$.
The functions nag_rand_field_2d_user_setup (g05zqc) and nag_rand_field_2d_generate (g05zsc) are used to simulate a two-dimensional stationary Gaussian random field, with mean function zero and variogram $\gamma \left(\mathbf{x}\right)$, over a domain $\left[{x}_{\mathrm{min}},{x}_{\mathrm{max}}\right]×\left[{y}_{\mathrm{min}},{y}_{\mathrm{max}}\right]$, using an equally spaced set of ${N}_{1}×{N}_{2}$ points; ${N}_{1}$ points in the $x$-direction and ${N}_{2}$ points in the $y$-direction. The problem reduces to sampling a Normal random vector $\mathbf{X}$ of size ${N}_{1}×{N}_{2}$, with mean vector zero and a symmetric covariance matrix $A$, which is an ${N}_{2}$ by ${N}_{2}$ block Toeplitz matrix with Toeplitz blocks of size ${N}_{1}$ by ${N}_{1}$. Since $A$ is in general expensive to factorize, a technique known as the circulant embedding method is used. $A$ is embedded into a larger, symmetric matrix $B$, which is an ${M}_{2}$ by ${M}_{2}$ block circulant matrix with circulant blocks of size ${M}_{1}$ by ${M}_{1}$, where ${M}_{1}\ge 2\left({N}_{1}-1\right)$ and ${M}_{2}\ge 2\left({N}_{2}-1\right)$. $B$ can now be factorized as $B=W\Lambda {W}^{*}={R}^{*}R$, where $W$ is the two-dimensional Fourier matrix (${W}^{*}$ is the complex conjugate of $W$), $\Lambda$ is the diagonal matrix containing the eigenvalues of $B$ and $R={\Lambda }^{\frac{1}{2}}{W}^{*}$. $B$ is known as the embedding matrix. The eigenvalues can be calculated by performing a discrete Fourier transform of the first row (or column) of $B$ and multiplying by ${M}_{1}×{M}_{2}$, and so only the first row (or column) of $B$ is needed – the whole matrix does not need to be formed.
The symmetry of $A$ as a block matrix, and the symmetry of each block of $A$, depends on whether the variogram $\gamma$ is even or not. $\gamma$ is even in its first coordinate if $\gamma \left({\left[{-x}_{1},{x}_{2}\right]}^{\mathrm{T}}\right)=\gamma \left({\left[{x}_{1},{x}_{2}\right]}^{\mathrm{T}}\right)$, even in its second coordinate if $\gamma \left({\left[{x}_{1},{-x}_{2}\right]}^{\mathrm{T}}\right)=\gamma \left({\left[{x}_{1},{x}_{2}\right]}^{\mathrm{T}}\right)$, and even if it is even in both coordinates (in two dimensions it is impossible for $\gamma$ to be even in one coordinate and uneven in the other). If $\gamma$ is even then $A$ is a symmetric block matrix and has symmetric blocks; if $\gamma$ is uneven then $A$ is not a symmetric block matrix and has non-symmetric blocks. In the uneven case, ${M}_{1}$ and ${M}_{2}$ are set to be odd in order to guarantee symmetry in $B$.
As long as all of the values of $\Lambda$ are non-negative (i.e., $B$ is positive semidefinite), $B$ is a covariance matrix for a random vector $\mathbf{Y}$ which has ${M}_{2}$ blocks of size ${M}_{1}$. Two samples of $\mathbf{Y}$ can now be simulated from the real and imaginary parts of ${R}^{*}\left(\mathbf{U}+i\mathbf{V}\right)$, where $\mathbf{U}$ and $\mathbf{V}$ have elements from the standard Normal distribution. Since ${R}^{*}\left(\mathbf{U}+i\mathbf{V}\right)=W{\Lambda }^{\frac{1}{2}}\left(\mathbf{U}+i\mathbf{V}\right)$, this calculation can be done using a discrete Fourier transform of the vector ${\Lambda }^{\frac{1}{2}}\left(\mathbf{U}+i\mathbf{V}\right)$. Two samples of the random vector $\mathbf{X}$ can now be recovered by taking the first ${N}_{1}$ elements of the first ${N}_{2}$ blocks of each sample of $\mathbf{Y}$ – because the original covariance matrix $A$ is embedded in $B$, $\mathbf{X}$ will have the correct distribution.
If $B$ is not positive semidefinite, larger embedding matrices $B$ can be tried; however if the size of the matrix would have to be larger than maxm, an approximation procedure is used. We write $\Lambda ={\Lambda }_{+}+{\Lambda }_{-}$, where ${\Lambda }_{+}$ and ${\Lambda }_{-}$ contain the non-negative and negative eigenvalues of $B$ respectively. Then $B$ is replaced by $\rho {B}_{+}$ where ${B}_{+}=W{\Lambda }_{+}{W}^{*}$ and $\rho \in \left(0,1\right]$ is a scaling factor. The error $\epsilon$ in approximating the distribution of the random field is given by
 $ε= 1-ρ 2 trace⁡Λ + ρ2 trace⁡Λ- M .$
Three choices for $\rho$ are available, and are determined by the input argument corr:
• setting ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleTraces}$ sets
 $ρ= trace⁡Λ trace⁡Λ+ ,$
• setting ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleSqrtTraces}$ sets
 $ρ= trace⁡Λ trace⁡Λ+ ,$
• setting ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleOne}$ sets $\rho =1$.
nag_rand_field_2d_user_setup (g05zqc) finds a suitable positive semidefinite embedding matrix $B$ and outputs its sizes in the vector m and the square roots of its eigenvalues in lam. If approximation is used, information regarding the accuracy of the approximation is output. Note that only the first row (or column) of $B$ is actually formed and stored.

## 4  References

Dietrich C R and Newsam G N (1997) Fast and exact simulation of stationary Gaussian processes through circulant embedding of the covariance matrix SIAM J. Sci. Comput. 18 1088–1107
Schlather M (1999) Introduction to positive definite functions and to unconditional simulation of random fields Technical Report ST 99–10 Lancaster University
Wood A T A and Chan G (1994) Simulation of stationary Gaussian processes in ${\left[0,1\right]}^{d}$ Journal of Computational and Graphical Statistics 3(4) 409–432

## 5  Arguments

1:     ns[$2$]const IntegerInput
On entry: the number of sample points to use in each direction, with ${\mathbf{ns}}\left[0\right]$ sample points in the $x$-direction, ${N}_{1}$ and ${\mathbf{ns}}\left[1\right]$ sample points in the $y$-direction, ${N}_{2}$. The total number of sample points on the grid is therefore ${\mathbf{ns}}\left[0\right]×{\mathbf{ns}}\left[1\right]$.
Constraints:
• ${\mathbf{ns}}\left[0\right]\ge 1$;
• ${\mathbf{ns}}\left[1\right]\ge 1$.
2:     xmindoubleInput
On entry: the lower bound for the $x$-coordinate, for the region in which the random field is to be simulated.
Constraint: ${\mathbf{xmin}}<{\mathbf{xmax}}$.
3:     xmaxdoubleInput
On entry: the upper bound for the $x$-coordinate, for the region in which the random field is to be simulated.
Constraint: ${\mathbf{xmin}}<{\mathbf{xmax}}$.
4:     ymindoubleInput
On entry: the lower bound for the $y$-coordinate, for the region in which the random field is to be simulated.
Constraint: ${\mathbf{ymin}}<{\mathbf{ymax}}$.
5:     ymaxdoubleInput
On entry: the upper bound for the $y$-coordinate, for the region in which the random field is to be simulated.
Constraint: ${\mathbf{ymin}}<{\mathbf{ymax}}$.
6:     maxm[$2$]const IntegerInput
On entry: determines the maximum size of the circulant matrix to use – a maximum of ${\mathbf{maxm}}\left[0\right]$ elements in the $x$-direction, and a maximum of ${\mathbf{maxm}}\left[1\right]$ elements in the $y$-direction. The maximum size of the circulant matrix is thus ${\mathbf{maxm}}\left[0\right]$$×$${\mathbf{maxm}}\left[1\right]$.
Constraints:
• if ${\mathbf{parity}}=\mathrm{Nag_Even}$, ${\mathbf{maxm}}\left[i\right]\ge {2}^{k}$, where $k$ is the smallest integer satisfying ${2}^{k}\ge 2\left({\mathbf{ns}}\left[i\right]-1\right)$, for $i=0,1$ ;
• if ${\mathbf{parity}}=\mathrm{Nag_Odd}$, ${\mathbf{maxm}}\left[i\right]\ge {3}^{k}$, where $k$ is the smallest integer satisfying ${3}^{k}\ge 2\left({\mathbf{ns}}\left[i\right]-1\right)$, for $i=0,1$ .
7:     vardoubleInput
On entry: the multiplicative factor ${\sigma }^{2}$ of the variogram $\gamma \left(\mathbf{x}\right)$.
Constraint: ${\mathbf{var}}\ge 0.0$.
8:     cov2function, supplied by the userExternal Function
cov2 must evaluate the variogram $\gamma \left(\mathbf{x}\right)$ for all $\mathbf{x}$ if ${\mathbf{parity}}=\mathrm{Nag_Odd}$, and for all $\mathbf{x}$ with non-negative entries if ${\mathbf{parity}}=\mathrm{Nag_Even}$. The value returned in gamma is multiplied internally by var.
The specification of cov2 is:
 void cov2 (double x, double y, double *gamma, Nag_Comm *comm)
1:     xdoubleInput
On entry: the coordinate $x$ at which the variogram $\gamma \left(\mathbf{x}\right)$ is to be evaluated.
2:     ydoubleInput
On entry: the coordinate $y$ at which the variogram $\gamma \left(\mathbf{x}\right)$ is to be evaluated.
On exit: the value of the variogram $\gamma \left(\mathbf{x}\right)$.
4:     commNag_Comm *
Pointer to structure of type Nag_Comm; the following members are relevant to cov2.
userdouble *
iuserInteger *
pPointer
The type Pointer will be void *. Before calling nag_rand_field_2d_user_setup (g05zqc) you may allocate memory and initialize these pointers with various quantities for use by cov2 when called from nag_rand_field_2d_user_setup (g05zqc) (see Section 3.2.1.1 in the Essential Introduction).
9:     parityNag_ParityInput
On entry: indicates whether the covariance function supplied is even or uneven.
${\mathbf{parity}}=\mathrm{Nag_Odd}$
The covariance function is uneven.
${\mathbf{parity}}=\mathrm{Nag_Even}$
The covariance function is even.
Constraint: ${\mathbf{parity}}=\mathrm{Nag_Odd}$ or $\mathrm{Nag_Even}$.
On entry: determines whether the embedding matrix is padded with zeros, or padded with values of the variogram. The choice of padding may affect how big the embedding matrix must be in order to be positive semidefinite.
${\mathbf{pad}}=\mathrm{Nag_EmbedPadZeros}$
The embedding matrix is padded with zeros.
${\mathbf{pad}}=\mathrm{Nag_EmbedPadValues}$
The embedding matrix is padded with values of the variogram.
Suggested value: ${\mathbf{pad}}=\mathrm{Nag_EmbedPadValues}$.
Constraint: ${\mathbf{pad}}=\mathrm{Nag_EmbedPadZeros}$ or $\mathrm{Nag_EmbedPadValues}$.
11:   corrNag_EmbedScaleInput
On entry: determines which approximation to implement if required, as described in Section 3.
Suggested value: ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleTraces}$.
Constraint: ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleTraces}$, $\mathrm{Nag_EmbedScaleSqrtTraces}$ or $\mathrm{Nag_EmbedScaleOne}$.
12:   lam[${\mathbf{maxm}}\left[0\right]×{\mathbf{maxm}}\left[1\right]$]doubleOutput
On exit: contains the square roots of the eigenvalues of the embedding matrix.
13:   xx[${\mathbf{ns}}\left[0\right]$]doubleOutput
On exit: the points of the $x$-coordinates at which values of the random field will be output.
14:   yy[${\mathbf{ns}}\left[1\right]$]doubleOutput
On exit: the points of the $y$-coordinates at which values of the random field will be output.
15:   m[$2$]IntegerOutput
On exit: ${\mathbf{m}}\left[0\right]$ contains ${M}_{1}$, the size of the circulant blocks and ${\mathbf{m}}\left[1\right]$ contains ${M}_{2}$, the number of blocks, resulting in a final square matrix of size ${M}_{1}×{M}_{2}$.
16:   approxInteger *Output
On exit: indicates whether approximation was used.
${\mathbf{approx}}=0$
No approximation was used.
${\mathbf{approx}}=1$
Approximation was used.
17:   rhodouble *Output
On exit: indicates the scaling of the covariance matrix. ${\mathbf{rho}}=1.0$ unless approximation was used with ${\mathbf{corr}}=\mathrm{Nag_EmbedScaleTraces}$ or $\mathrm{Nag_EmbedScaleSqrtTraces}$.
18:   icountInteger *Output
On exit: indicates the number of negative eigenvalues in the embedding matrix which have had to be set to zero.
19:   eig[$3$]doubleOutput
On exit: indicates information about the negative eigenvalues in the embedding matrix which have had to be set to zero. ${\mathbf{eig}}\left[0\right]$ contains the smallest eigenvalue, ${\mathbf{eig}}\left[1\right]$ contains the sum of the squares of the negative eigenvalues, and ${\mathbf{eig}}\left[2\right]$ contains the sum of the absolute values of the negative eigenvalues.
20:   commNag_Comm *Communication Structure
The NAG communication argument (see Section 3.2.1.1 in the Essential Introduction).
21:   failNagError *Input/Output
The NAG error argument (see Section 3.6 in the Essential Introduction).

## 6  Error Indicators and Warnings

On entry, argument $⟨\mathit{\text{value}}⟩$ had an illegal value.
NE_INT_ARRAY
On entry, ${\mathbf{maxm}}=\left[⟨\mathit{\text{value}}⟩,⟨\mathit{\text{value}}⟩\right]$.
Constraint: the minima for maxm are $\left[⟨\mathit{\text{value}}⟩,⟨\mathit{\text{value}}⟩\right]$.
Where, if ${\mathbf{parity}}=\mathrm{Nag_Even}$, the minimum calculated value of ${\mathbf{maxm}}\left[\mathit{i}-1\right]$ is given by ${2}^{k}$, where $k$ is the smallest integer satisfying ${2}^{k}\ge 2\left({\mathbf{ns}}\left[\mathit{i}-1\right]-1\right)$, and if ${\mathbf{parity}}=\mathrm{Nag_Odd}$, the minimum calculated value of ${\mathbf{maxm}}\left[\mathit{i}-1\right]$ is given by ${3}^{k}$, where $k$ is the smallest integer satisfying ${3}^{k}\ge 2\left({\mathbf{ns}}\left[\mathit{i}-1\right]-1\right)$, for $\mathit{i}=1,2$.
On entry, ${\mathbf{ns}}=\left[⟨\mathit{\text{value}}⟩,⟨\mathit{\text{value}}⟩\right]$.
Constraint: ${\mathbf{ns}}\left[0\right]\ge 1$, ${\mathbf{ns}}\left[1\right]\ge 1$.
NE_INTERNAL_ERROR
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
NE_REAL
On entry, ${\mathbf{var}}=⟨\mathit{\text{value}}⟩$.
Constraint: ${\mathbf{var}}\ge 0.0$.
NE_REAL_2
On entry, ${\mathbf{xmin}}=⟨\mathit{\text{value}}⟩$ and ${\mathbf{xmax}}=⟨\mathit{\text{value}}⟩$.
Constraint: ${\mathbf{xmin}}<{\mathbf{xmax}}$.
On entry, ${\mathbf{ymin}}=⟨\mathit{\text{value}}⟩$ and ${\mathbf{ymax}}=⟨\mathit{\text{value}}⟩$.
Constraint: ${\mathbf{ymin}}<{\mathbf{ymax}}$.

## 7  Accuracy

If on exit ${\mathbf{approx}}=1$, see the comments in Section 3 regarding the quality of approximation; increase the values in maxm to attempt to avoid approximation.

## 8  Parallelism and Performance

nag_rand_field_2d_user_setup (g05zqc) is not threaded by NAG in any implementation.
nag_rand_field_2d_user_setup (g05zqc) makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the Users' Note for your implementation for any additional implementation-specific information.

None.

## 10  Example

This example calls nag_rand_field_2d_user_setup (g05zqc) to calculate the eigenvalues of the embedding matrix for $25$ sample points on a $5$ by $5$ grid of a two-dimensional random field characterized by the symmetric stable variogram:
 $γx = σ2 exp - x′ ν ,$
where ${x}^{\prime }=\left|\frac{x}{{\ell }_{1}}+\frac{y}{{\ell }_{2}}\right|$, and ${\ell }_{1}$, ${\ell }_{2}$ and $\nu$ are parameters.
It should be noted that the symmetric stable variogram is one of the pre-defined variograms available in nag_rand_field_2d_predef_setup (g05zrc). It is used here purely for illustrative purposes.

### 10.1  Program Text

Program Text (g05zqce.c)

### 10.2  Program Data

Program Data (g05zqce.d)

### 10.3  Program Results

Program Results (g05zqce.r)