Hello I am trying to use predict from an arma-Garch model (arma(2, 2) +
garch(1, 1)) and I am getting the following error:
Error en arima(x = object@data, order = c(max(u, 1), 0, max(v, 1)), init c(ar,
:
non-stationary AR part from CSS
Does anybody know what can be the reason of this error? The model I have
estimated is the following:
Title:
GARCH Modelling
Call:
garchFit(formula = fmla, data = X[, i], trace = F)
Mean and Variance Equation:
data ~ arma(2, 2) + garch(1, 1)
<environment: 060ec530>
[data = X[, i]]
Conditional Distribution:
norm
Coefficient(s):
mu ar1 ar2 ma1 ma2 omega
alpha1 beta1
-0.00079014 0.44934211 0.64374977 -0.46541382 -0.23879607 0.00028933
0.52056245 0.00000001
Std. Errors:
based on Hessian
Error Analysis:
Estimate Std. Error t value Pr(>|t|)
mu -7.901e-04 5.649e-04 -1.399 0.16187
ar1 4.493e-01 1.498e-01 3.000 0.00270 **
ar2 6.437e-01 1.453e-01 4.429 9.46e-06 ***
ma1 -4.654e-01 1.494e-01 -3.115 0.00184 **
ma2 -2.388e-01 1.318e-01 -1.812 0.07006 .
omega 2.893e-04 3.899e-05 7.420 1.17e-13 ***
alpha1 5.206e-01 NA NA NA
beta1 1.000e-08 NA NA NA
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Log Likelihood:
297.5401 normalized: 2.438853
Thank you
Felipe Parra
[[alternative HTML version deleted]]
