Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Regressions for growth rates
Message posted by Russell D. Pritchard on November 21, 1999 at 12:00 AM (ET)
With respect to performing regression analysis on growth rates, could someone tell me what the "seemingly unrelated" (SUR)technique is?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Regressions for growth rates
Message posted by JG on November 21, 1999 at 12:00 AM (ET)
I do not know what 'seemingly unrelated' refers to. However, remember that growth rates implies that you are doing a logarithmic transformation to make the problem suitable for linear regression. What that does to collinearity I do not know. In any case, even when you do your linesr regression successfully the results have to be transformed into the original model and interpreted correctly in terms of the original model. A book of statistics for chemical engineers will probably discuss this type of regression.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.