[R] How can I get estimates including standard errors for parameters in constraint for retricted models

[bonnychen]

Hello,

I am running linear regressions on a restricted model, in which one
parameter, say gamma1,  equals alpha/(1-alpha), and another parameter, say
gamma2, equals beta/(1-alpha), i.e.:

y = c0 + gamma1*x1 + gamma2*x2,

where gamma1= alpha/(1-alpha), gamma2=beta/(1-alpha).


I can get estimates of alpha and beta by solving equations given parameter
estimates of gamma1 and gamma2. However I notice that researchers also
report SE(standard errors) for alpha and beta in the literature, which are
different from one another, also different from the SE of gamma1 and gamma2. 

I don't know how to get SE for alpha and beta in this case. Nor do I know
whether there is some way so that I can get estimates along with SE for
alpha and beta from the regression on this restricted model, without solving
the system of equations separately. 

Can anyone kindly offer me some help on this question?

Many thanks,
Bonny
-- 
View this message in context: http://www.nabble.com/How-can-I-get-estimates-including-standard-errors-for-parameters-in-constraint-for-retricted-models-tp24253359p24253359.html
Sent from the R help mailing list archive at Nabble.com.