Dear S-Plus users,
I need some help in solving the problem using the S-Plus software. Sorry to
say that I am new to this software.
Here I have a non-linear model with additive terms yi =
â0*(exp(-â0*xi)-exp(-â1*xi))/ (â0-â1) + åi where åi are independent and
identically distributed random variables with mean 0 and unknown variance
ó2.The problem is I need to use these four techniques to estimate the betas
, get the MSE(Mean Squared Error) and list the residuals for the model
Technique 1: linearize
Technique 2: non-linear least squares estimate
Technique 3: Using Lawrence and Arthur (1990) proposal of one-step iterative
reweighting scheme in which a weighted non-linear least squares process is
performed on the model using Huber's weight function wi defined as 1 ,if |
ei | < c and c/ei ,if | ei | > c where c is the tuning constant and ei is
the standardized absolute residual. ei = | ri | / s .Lawrence and Arthur
(1990) defined s as the standard deviation of the residuals {ri ; i = 1, ..,
n} obtained from the least-squares procedure
Technique 4: reweighted linear jackknife (RLQ)
Could any expertise out there please advise me on how to use the S-Plus by
clicking the Menus and Dialogs step by steps if possible or otherwise by
S-Plus syntax and grammar to write a program?
Thank you in advance for spending your expensive time in giving me the
advice. All the help given is much appreciated .
Wishing all the users out the MERRY CHRISTMAS AND A HAPPY NEW YEAR.
Thank you.
From chan MALI CHAN.
_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.
|