[R-sig-eco] How to handle Inf values in VGAM
Rajendra Mohan Panda
rmp.iit.kgp at gmail.com
Sun Oct 11 09:26:40 CEST 2015
Dear All
I thank you all for your response and making this forum a 'knowledge hub'.
I have immensely benefited from this forum scientists. This time the error
is on handling infinite values in VGAM model run. Because of infinite
values, I get following errors:
VGAM s.vam loop 1 : loglikelihood = -6516.0523
VGAM s.vam loop 2 : loglikelihood = -Inf
Error in s.vam(x, z, wz, tfit$smomat, which, tfit$smooth.frame, bf.maxit,
:
NA/NaN/Inf in foreign function call (arg 4)
I use following formula for my model:
vgam(cbind(species.list) ~s(hanpp), data = SPEV, binomialff(link = "logit",
multiple.responses = TRUE, bred = TRUE), maxit = 100,trace = TRUE)
N.B. Species data format includes only 0 or 1 with no NA or missing values
Kindly advise if there is any way to check this error prior to model run
or during the modelling process.
Your kind response is highly appreciated
Best Regards
Rajendra M Panda
School of Water Resources
Indian Institute of Technology Kharagpur
[[alternative HTML version deleted]]
More information about the R-sig-ecology
mailing list