Hi R-users,
I have 1020 time series ( each of length 10,000), say, X1,X2,......,X1020
and I want to perform Factor Analysis using 50 factors on their correlation
matrix.
The issue is: for every series, I have a weight, i.e. *the series X_i has a
pre-defined weight of w_i* ( i = 1,2,...., 1020). I want to estimate the
factor loadings and specific variances in the model by optimizing the
likelihood function (assuming multivariate normality, as usual).
Is it possible to estimate the model parameters using the weights for each
time series variable in the objective function?
One comment here - For computational purposes or otherwise, it is *ok to
change my objective function* (instead of taking the likelihood function,
may be something like minimizing the weighted sum of squared specific
variances for the variables would make sense).
Any help with this will be really appreciated.
Regards,
Preetam
--
Preetam Pal
(+91)-9432212774
M-Stat 2nd Year, Room No. N-114
Statistics Division, C.V.Raman
Hall
Indian Statistical Institute, B.H.O.S.
Kolkata.
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
