Hi all,
I'm running into a strange problem running Amelia. I receive the following
error when I try to run Amelia with about 60 variable (N = 2500):
Error in amcheck(x = x, m = m, idvars = numopts$idvars, priors = priors, :
It then lists about half of my variables as being "perfectly collinear".
This is improbable at best. Besides trimming my dataset starting with all
problematic variables, then slowly adding them back in, are there any other
potential solutions?
Thanks!
Sara Anderson
Original code:
amelia(x = getAmelia("amelia.data"), m = 5, idvars = c("studentid",
"bg_id"), ts = NULL, cs = NULL, priors = NULL, lags = NULL, empri = 0,
intercs = FALSE, leads = NULL, splinetime = NULL, logs = NULL,
sqrts = NULL, lgstc = NULL, ords = NULL, noms = c("gifted_14",
"ell_14", "sped_14", "mom_conf", "peer_1d", "peer_2d", "grade_14",
"district", "female", "ps", "lastyrpk", "lastyrhs", "lwfather",
"internet", "race2", "tpslunch0607", "daycare_oh3", "preschool3",
"headstart3", "centerbased3", "foreignborn", "english", "momeduc",
"sus_13d", "marital_r"), bounds = c(1, 2, 3, 6, 7, 11, 14,
42, 43, 0, 0, 0, 0, 0, 0, 0, 400, 400, 4, 4, 4, 990, 990,
4, 2, 990, 990), max.resample = 1000, tolerance = 1e-04)
I would like to produce log-log plots from overimpute. If I try adding a
graphical parameter, log='y' I get an error
Error in plot.window(...) : Logarithmic axis must have positive limits
My data are all positive.
Looking at the code for overimpute, there is a variable
addedroom <- (max(uppers) - min(lowers)) * 0.1
which is then used as part of the ylim argument for the plot
ylim = range(c(lowers - addedroom, uppers))
I think this is making the y-axis go slightly into negative territory so
the log doesn't work.
Can anyone suggest a work around?
--
E: tony.ladson(a)moroka.com.au
M: 0408 440 028
A: P.O. Box 1245, Fitzroy North, Vic 3068
B: http://tonyladson.wordpress.com/