need help on GARCH function

[Richard <hujian@gmail.com>]

Hi Guys,

 

I am using S+6.1 (with Finmetrics) to model financial return series, the code is like:

 

>CHN.margin=garch(DS.ret[,1:7]~arma(1, 0),~garch(1, 1))

 

where DS.ret [,1:7] is seven series (col 1 through 7), arma(1,0) means AR(1) in mean equation, garch(1,1) means p=1 and q=1 in variance equation.

 

I got the following result:

 

Iteration   0  Step Size =  1.00000  Likelihood = -1.60285
Iteration   0  Step Size =  2.00000  Likelihood = -1.00000e+010
Iteration   1  Step Size =  1.00000  Likelihood = -1.59993
Iteration   1  Step Size =  2.00000  Likelihood = -1.59727
Iteration   1  Step Size =  4.00000  Likelihood = -1.59471
Iteration   1  Step Size =  8.00000  Likelihood = -1.00000e+010
Iteration   2  Step Size =  1.00000  Likelihood = -1.59350
Iteration   2  Step Size =  2.00000  Likelihood = -1.59290
Iteration   2  Step Size =  4.00000  Likelihood = -1.59349
Iteration   3  Step Size =  1.00000  Likelihood = -1.59231
Iteration   3  Step Size =  2.00000  Likelihood = -1.59196
Iteration   3  Step Size =  4.00000  Likelihood = -1.59200
Iteration   4  Step Size =  1.00000  Likelihood = -1.59160
Iteration   4  Step Size =  2.00000  Likelihood = -1.59139
Iteration   4  Step Size =  4.00000  Likelihood = -1.59142
Iteration   5  Step Size =  1.00000  Likelihood = -1.59117
Iteration   5  Step Size =  2.00000  Likelihood = -1.59104
Iteration   5  Step Size =  4.00000   Likelihood = -1.59109
Iteration   6  Step Size =  1.00000  Likelihood = -1.59089
Iteration   6  Step Size =  2.00000  Likelihood = -1.59081
Iteration   6  Step Size =  4.00000  Likelihood = -1.59087
Iteration   7  Step Size =  1.00000  Likelihood = -1.59072
Iteration   7  Step Size =  2.00000  Likelihood = -1.59067
Iteration   7  Step Size =  4.00000  Likelihood = -1.59073

Convergence R-Square = 0.00008684686 is less than tolerance = 0.0001
Convergence reached.
Problem in x[i,  , drop = F]: Array subscript (30968) out of bounds, should be at most 4424
Use traceback() to see the call stack

When I used traceback, I got the following message:

 

> traceback()
23: eval(action, sys.parent())
22: doErrorAction("Problem in x[i,  , drop = F]: Array subscrip
t (30968) out of bounds, should be at most 4424",
21: x[i,  , drop = F]
20: sub( x@data, i,  )
19: is(x, "character") || is(x, "numeric")
18: is(x, "character") || is(x, "numeric") || is(x, "complex")
17: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
16: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
15: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
14: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
13: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
12: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
11: is(x, "character") || is(x, "numeric") || is(x, "complex")
||
10: is.rectangular(x)
9: as.rectangular(x)
8: asSeriesData(sub(x@data, i,  ))
7: seriesOrig[series.start:Ty]
6: garch(DS.ret[, 1:7] ~ arma(1, 0),  ~ garch(1, 1))
5: eval(i, local)
4: source(auto.print = auto.print, exprs = substitute(
3: script.run(exprs.literal = {
2: eval(_expression_(script.run(exprs.literal = {
1:
Message: Problem in x[i,  , drop = F]: Array subscript (30968)
out of bounds, should be at most 4424
>

 

I don't know what is wrong here. Any thoughts? Thanks in advance!!

 

Richard