title error in mgarch's garch output

[maxchen]

ver: EViews7Patch_060211.exe

try the following example

Code: Select all

%ex7 = "\Example Files\EV7 Manual Data\"
%ch31 = "Chapter 31 - System Estimation"
%wf = @evpath + %ex7 +%ch31 +"\fx.wf1"
wfopen %wf

system sys01
sys01.append dlog(jy) = c(1)
sys01.append dlog(sf) = c(2)
sys01.append dlog(bp) = c(3)

smpl 1980 2000
freeze(tbdv) sys01.arch(deriv=aa,showopts) @diagvech  c(indef)   arch(1,indef)   garch(1,indef)

freeze(g_cor) sys01.garch(cor)   'correlation typo on the output fig as covariance


the last line, the output figure titled as "conditional covariance", not correlation. however, using point and click, it is fine.

[EViews Gareth]

Will be fixed in next patch.

[maxchen]

Ver: EViews7Patch_061311.exe, the title bug has been fixed, however for asymmetry term, there may be a typo in the output and representations in Covariance specification part

Code: Select all

freeze(tb01) sys01.arch @diagvech  c(fullrank)   arch(1,fullrank)   garch(1,fullrank) tarch(1,fullrank)
freeze(txt01) sys01.representations


the Covariance specification says
GARCH = M + A1.*RESID(-1)*RESID(-1)' + D1.*(RESID(-1)*(RESID(-1)<0))
*(RESID(-1)*(RESID(-1)<0))'D1.*(RESID(-1)*(RESID(-1)<0))
*(RESID(-1)*(RESID(-1)<0))' + B1.*GARCH(-1)
it seems that the tarch term has been duplicated.

[maxchen]

another bug

Code: Select all

sys01.makeloglike slk
sys01.makeloglike slk2 'error if comment inline


the first line work, however, the second line errors, if the comment is removed, it is OK.

[maxchen]

one more

Code: Select all

sys01.correl(12,factor=cov)


will generate the same graph as sys01.correl(12)
if using point and click, it is fine.