g13 Chapter Contents
g13 Chapter Introduction
NAG C Library Manual

# NAG Library Function Documentnag_tsa_multi_diff (g13dlc)

## 1  Purpose

nag_tsa_multi_diff (g13dlc) differences and/or transforms a multivariate time series.

## 2  Specification

 #include #include
 void nag_tsa_multi_diff (Integer k, Integer n, const double z[], const Integer tr[], const Integer id[], const double delta[], double w[], Integer *nd, NagError *fail)

## 3  Description

For certain time series it may first be necessary to difference the original data to obtain a stationary series before calculating autocorrelations, etc. This function also allows you to apply either a square root or a log transformation to the original time series to stabilize the variance if required.
If the order of differencing required for the $i$th series is ${\mathit{d}}_{i}$, then the differencing operator is defined by ${\delta }_{i}\left(B\right)=1-{\delta }_{i1}B-{\delta }_{i2}{B}^{2}-\cdots -{\delta }_{i{\mathit{d}}_{i}}{B}^{{\mathit{d}}_{i}}$, where $B$ is the backward shift operator; that is, $B{Z}_{t}={Z}_{t-1}$. Let $\mathit{d}$ denote the maximum of the orders of differencing, ${\mathit{d}}_{i}$, over the $k$ series. The function computes values of the differenced/transformed series ${W}_{\mathit{t}}={\left({w}_{1\mathit{t}},{w}_{2\mathit{t}},\dots ,{w}_{\mathit{k}\mathit{t}}\right)}^{\mathrm{T}}$, for $\mathit{t}=\mathit{d}+1,\dots ,n$, as follows:
 $wit=δiBzit*, i=1,2,…,k$
where ${z}_{it}^{*}$ are the transformed values of the original $k$-dimensional time series ${Z}_{t}={\left({z}_{1t},{z}_{2t},\dots ,{z}_{kt}\right)}^{\mathrm{T}}$.
The differencing parameters ${\delta }_{ij}$, for $i=1,2,\dots ,k$ and $j=1,2,\dots ,{\mathit{d}}_{i}$, must be supplied by you. If the $i$th series does not require differencing, then ${\mathit{d}}_{i}=0$.

## 4  References

Box G E P and Jenkins G M (1976) Time Series Analysis: Forecasting and Control (Revised Edition) Holden–Day
Wei W W S (1990) Time Series Analysis: Univariate and Multivariate Methods Addison–Wesley

## 5  Arguments

1:     kIntegerInput
On entry: $k$, the dimension of the multivariate time series.
Constraint: ${\mathbf{k}}\ge 1$.
2:     nIntegerInput
On entry: $n$, the number of observations in the series, prior to differencing.
Constraint: ${\mathbf{n}}\ge 1$.
3:     z[${\mathbf{k}}×{\mathbf{n}}$]const doubleInput
On entry: ${\mathbf{z}}\left[\left(\mathit{t}-1\right)k+\mathit{i}-1\right]$ must contain the $\mathit{i}$th series at time $\mathit{t}$, for $\mathit{t}=1,2,\dots ,n$ and $\mathit{i}=1,2,\dots ,k$.
4:     tr[k]const IntegerInput
On entry: ${\mathbf{tr}}\left[\mathit{i}-1\right]$ indicates whether the $\mathit{i}$th series is to be transformed, for $\mathit{i}=1,2,\dots ,k$.
${\mathbf{tr}}\left[i-1\right]=-1$
A square root transformation is used.
${\mathbf{tr}}\left[i-1\right]=0$
No transformation is used.
${\mathbf{tr}}\left[i-1\right]=1$
A log transformation is used.
Constraint: ${\mathbf{tr}}\left[\mathit{i}-1\right]=-1$, $0$ or $1$, for $\mathit{i}=1,2,\dots ,k$.
5:     id[k]const IntegerInput
On entry: the order of differencing for each series, ${\mathit{d}}_{1},{\mathit{d}}_{2},\dots ,{\mathit{d}}_{k}$.
Constraint: $0\le {\mathbf{id}}\left[\mathit{i}\right]<{\mathbf{n}}$, for $\mathit{i}=0,1,\dots ,{\mathbf{k}}-1$.
6:     delta[$\mathit{dim}$]const doubleInput
Note: the dimension, dim, of the array delta must be at least ${\mathbf{k}}×\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,\mathit{d}\right)$, where $\mathit{d}=\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left({\mathbf{id}}\left[i-1\right]\right)$.
On entry: if ${\mathbf{id}}\left[\mathit{i}-1\right]>0$ then ${\mathbf{delta}}\left[\left(\mathit{j}-1\right)k+\mathit{i}-1\right]$ must be set to ${\delta }_{\mathit{i}\mathit{j}}$, for $\mathit{j}=1,2,\dots ,{\mathit{d}}_{l}$ and $\mathit{i}=1,2,\dots ,k$.
7:     w[$\mathit{dim}$]doubleOutput
Note: the dimension, dim, of the array w must be at least ${\mathbf{k}}×\left({\mathbf{n}}-\mathit{d}\right)$, where $\mathit{d}=\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left({\mathbf{id}}\left[i-1\right]\right)$.
On exit: ${\mathbf{w}}\left[\left(\mathit{t}-1\right)k+\mathit{i}-1\right]$ contains the value of ${w}_{\mathit{i},\mathit{t}+\mathit{d}}$, for $\mathit{i}=1,2,\dots ,k$ and $\mathit{t}=1,2,\dots ,n-\mathit{d}$.
8:     ndInteger *Output
On exit: the number of differenced values, $n-\mathit{d}$, in the series, where $\mathit{d}=\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left({\mathbf{id}}\left[i-1\right]\right)$.
9:     failNagError *Input/Output
The NAG error argument (see Section 3.6 in the Essential Introduction).

## 6  Error Indicators and Warnings

NE_ALLOC_FAIL
Dynamic memory allocation failed.
On entry, argument $〈\mathit{\text{value}}〉$ had an illegal value.
NE_INT
On entry, ${\mathbf{k}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{k}}\ge 1$.
On entry, ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{n}}\ge 1$.
NE_INT_ARRAY
On entry, ${\mathbf{k}}=〈\mathit{\text{value}}〉$, ${\mathbf{id}}\left[\mathit{i}\right]=〈\mathit{\text{value}}〉$ and ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: $0\le {\mathbf{id}}\left[\mathit{i}\right]<{\mathbf{n}}$, for $\mathit{i}=0,1,\dots ,{\mathbf{k}}-1$.
On entry, ${\mathbf{tr}}\left[\mathit{i}-1\right]=〈\mathit{\text{value}}〉$ and ${\mathbf{k}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{tr}}\left[\mathit{i}-1\right]=-1$, $0$ or $1$, for $\mathit{i}=1,2,\dots ,k$.
NE_INT_ARRAY_ELEM_CONS
On entry, element $〈\mathit{\text{value}}〉$ of id is greater than or equal to n.
On entry, element $〈\mathit{\text{value}}〉$ of id is less than zero.
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_TRANSFORMATION
On entry, one (or more) of the transformations requested is invalid.

## 7  Accuracy

The computations are believed to be stable.

## 8  Further Comments

The same differencing operator does not have to be applied to all the series. For example, suppose we have $k=2$, and wish to apply the second-order differencing operator ${\nabla }^{2}$ to the first series and the first-order differencing operator $\nabla$ to the second series:
 $w1t =∇2z1t= 1-B 2z1t=1-2B+B2z1t, and w2t =∇z2t=1-Bz2t.$
Then ${\mathit{d}}_{1}=2,{\mathit{d}}_{2}=1$, $\mathit{d}=\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left({\mathit{d}}_{1},{\mathit{d}}_{2}\right)=2$, and
 $delta = δ11 δ12 δ21 = 2 -1 1 .$

## 9  Example

A program to difference (non-seasonally) each of two time series of length $48$. No transformation is to be applied to either of the series.

### 9.1  Program Text

Program Text (g13dlce.c)

### 9.2  Program Data

Program Data (g13dlce.d)

### 9.3  Program Results

Program Results (g13dlce.r)