Nick and Philippe,
Thank you for your advice.
And I'm sorry for my late reply.
With -mkmat- and sample codes Philippe gave, I could
get desired results.
> A chi-square test would require some programming on
> your part given a matrix data structure.
As to the chi-square test, the motivation is that
I wanted to test for homogeneity of two transition
probability matrices (created from the original ones
by normalization) across periods.
If it requires, as Nick commented, additional programming,
it might be better to look for some other strategies.
P.S.
BTW, what does 'MET' in 'Shorrocks' MET' stand for?
Though I read Shorrocks' Econometrica paper, I couldn't
come up with its original.
K.I.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
