NAG Library Function Document

nag_forecast_agarchI (g13fbc)

▸▿ Contents

    1  Purpose
    7  Accuracy
    10  Example

1
Purpose

nag_forecast_agarchI (g13fbc) forecasts the conditional variances, h t , t = 1 , … , τ  from a type I AGARCH p,q  sequence, where τ  is the forecast horizon (see Engle and Ng (1993)).

2
Specification

#include <nag.h>
#include <nagg13.h>
void  nag_forecast_agarchI (Integer num, Integer nt, Integer p, Integer q, const double theta[], double gamma, double fht[], const double ht[], const double et[], NagError *fail)

3
Description

Assume the standard γ=0  GARCH p,q  process can be represented by:
ε t ∣ ψ t-1 ∼ N 0, h t  
h t = α 0 + ∑ i=1 q α i ε t-i 2 + ∑ i=1 p β i h t-i ,   t = 1 , … , T .  
or type I AGARCH p,q  process with conditional variance h t  given by:
h t = α 0 + ∑ i=1 q α i ε t-i + γ 2 + ∑ i=1 p β i h t-i ,   t = 1 , … , T .  
has been modelled by nag_estimate_agarchI (g13fac) and the estimated conditional variances and residuals are contained in the arrays ht and et respectively. Then nag_forecast_agarchI (g13fbc) will use the last maxp,q  elements of the arrays ht and et to estimate the conditional variance forecasts, h t ∣ ψ T , where t = T + 1 , … , T + τ  and τ  is the forecast horizon.

4
References

Bollerslev T (1986) Generalised autoregressive conditional heteroskedasticity Journal of Econometrics 31 307–327
Engle R (1982) Autoregressive conditional heteroskedasticity with estimates of the variance of United Kingdom inflation Econometrica 50 987–1008
Engle R and Ng V (1993) Measuring and testing the impact of news on volatility Journal of Finance 48 1749–1777
Hamilton J (1994) Time Series Analysis Princeton University Press

5
Arguments

1:     num – IntegerInput
On entry: the number of terms in the arrays ht and et from the modelled sequence.
Constraint: maxp,q ≤ num .
2:     nt – IntegerInput
On entry: τ , the forecast horizon.
Constraint: nt>0 .
3:     p – IntegerInput
On entry: the GARCH p,q  argument p .
Constraint: 0 < maxp,q ≤ num , ​ p≥0 .
4:     q – IntegerInput
On entry: the GARCH p,q  argument q .
Constraint: 0 < maxp,q ≤ num , ​ q≥1 .
5:     theta[q+p+1] – const doubleInput
On entry: the first element must contain the coefficient α o  and the next q elements must contain the coefficients α i , for i=1,2,…,q. The remaining p elements must contain the coefficients β j , for j=1,2,…,p.
6:     gamma – doubleInput
On entry: the asymmetry argument γ  for the GARCH p,q  sequence.
7:     fht[nt] – doubleOutput
On exit: the forecast values of the conditional variance, h t , for t=1,2,…,τ.
8:     ht[num] – const doubleInput
On entry: the sequence of past conditional variances for the GARCH p,q  process, h t , for t=1,2,…,T.
9:     et[num] – const doubleInput
On entry: the sequence of past residuals for the GARCH p,q  process, ε t , for t=1,2,…,T.
10:   fail – NagError *Input/Output
The NAG error argument (see Section 3.7 in How to Use the NAG Library and its Documentation).

6
Error Indicators and Warnings

NE_2_INT_ARG_LT
On entry, num=value  while maxp,q = value. These arguments must satisfy num ≥ maxp,q .
NE_ALLOC_FAIL
Dynamic memory allocation failed.
NE_INT_ARG_LT
On entry, nt=value.
Constraint: nt≥1.
On entry, num=value.
Constraint: num≥0.
On entry, p=value.
Constraint: p≥0.
On entry, q=value.
Constraint: q≥1.

7
Accuracy

Not applicable.

8
Parallelism and Performance

nag_forecast_agarchI (g13fbc) is not threaded in any implementation.

9
Further Comments

None.

10
Example

See the example for nag_estimate_agarchI (g13fac).
© The Numerical Algorithms Group Ltd, Oxford, UK. 2017