[R] Weighted least squares with constraints

[GRANT Lewis]

Hi

I am attempting to convert my simple weighted regressions (produced using the weights argument in lm) to a constrained regression where the coefficients  sum to 1. I understand that I can do this using solve.qp and I have spent time reading the archives to understand how this is done, but I am unable to find an example of where the constraints were introduced in a weighted regression. 

I see that solve.qp will find the solution to min{(y-bx)^2} but can it be used for min{w((y-bx)^2)}, and how would I do this?

Thanks in advance

Lewis




 


**********************************************************************
Hermes Fund Managers Limited
Registered in England No. 1661776, 1 Portsoken Street, London E1 8HZ

***Please read the Hermes email disclaimer at http://www.hermes.co.uk/email_terms.htm before acting on this email or opening any attachment***
The contents of this email are confidential.  If you hav...{{dropped:17}}