literature for variance decomposition

[chuckychu]

Hello,

I need a textbook citation using variance decomposition for the following statement:

The estimates of the proportion of the forecast error variance are judged ‘‘significant’’ if the point estimate is at least twice as large as its standard error

or

if the point estimate is at least twice the estimated standard error it is considered to be significant

is there anyone that could help? Thanks in advance!

[Arun.stat]

I am not quite sure if I understood your question. Considering something as significant is completely subjective and any rule regarding that should be finalized before starting any experiment. Are you looking for something on what is the sampling distribution for the estimates corresponding to the Variance decomposition? Then try Lutkepohl: www.jmulti.de

Thanks,