|
| I am trying to do a GARCH analysis on foreign exchange rates. I
followed the example in the help file and it seemed fine. Now when I
try to import a different time series and do exactly what I did with
the previous one I obtain the following result:
===============================
Optimization terminated: magnitude of directional derivative in
search
direction less than 2*options.TolFun and maximum constraint
violation
is less than options.TolCon.
Active inequalities (to within options.TolCon = 1e-007):
lower upper ineqlin ineqnonlin
2 1
Boundary Constraints Active: Standard Errors May Be Inaccurate.
Mean: ARMAX(0,0,0); Variance: GARCH(1,1)
Conditional Probability Distribution: Gaussian
Number of Model Parameters Estimated: 4
Standard T
Parameter Value Error Statistic
----------- ----------- ------------ -----------
C 2.8937e-005 8.5699e-005 0.3377
K 2e-007 2.685e-008 7.4487
GARCH(1) 0.85703 0.011319 75.7160
ARCH(1) 0.14297 0.014785 9.6700
Log Likelihood Value: 4645.12
===============================
where A(1)+G(1) = 1 and it should be strictly less than zero. I use
the garchfit with returns not the raw data. What is wrong?
|
|