Hello
I do not think I have fully grasped how to use Amelia to deal with missing
data.
For instance, suppose I have a data.frame variable with 4 columns (year,
mon, ssn, dev) = (year, month, measurements, standard deviation of the
measurement). Of course, there are some random missing values on columns 3
and 4. The measurements are an almost periodic time-series contaminated by
noise.
I did some tests such as:
a.out <- amelia(x, m = 5, ts = "YEAR", p2s = 0, idvars=c("DEV"));
> a.out <- amelia(x, m = 5, ts = "YEAR", p2s = 0, idvars=c("MON","DEV"));
Error in matrix(1, AMn, 1) : non-numeric matrix extent
a.out <- amelia(x, m = 5, ts = "YEAR", p2s = 0);
a.out <- amelia(x, m = 5, p2s = 0, idvars=c("MON","DEV"));
If I got it right, idvars removes the columns from the imputation process.
Is that right?
There is an option called cs (cross-section) but I have no idea what it
does.
How can I use the information on column 4 to limit the values of the
estimates of the missing values? Some of the results are extremely high.
Any pointers that help me to understand what is going on will be most
appreciated.
Many thanks
Ed
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
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.
