Bivariate GARCH(1,1). Problem with "logl" - TIME-CRITICAL!!

[errdni]

Hello!
I followed the example program of bivariate GARCH (1,1), which I found after EVIews 7 installation.
Everything had been clear until Logl estimating process.

...
I wrote a command:
smpl s1
bvgarch.ml(showopts, m=100, c=1e-5)

And then I got the error:
"Missing values in @LOGL series at current coefficients at observation 2."

Can anybody explain as soon as possible, what should I do?

I attach the workfile that I had just before the command, I mentioned above.
And also full programm for my research.

[EViews Gareth]

http://forums.eviews.com/search.php?key ... mit=Search

[errdni]

I resolved this problem! I should have set the first observation of a sample "s0" a period after the first observation of a range of all the data; and, of course, then shift "s1".

But, anyway, thank you :)