Good afternoon everybody, I'm using optimization routines like optimix or nlminb to optimize the parameters of my dispersal curves which are "working" with different kernels :
/anlb2002z1b <- nlminb(c(0.5,2), objective=LogLiketot,lower=c(0,0), upper=c(1,200))/ where /LogLiketot/ contains my dispersal kernel parameters and /(0.5,2)/, an exemple of starting point to estimate these parameters. /anlb2002z1b/ returns me different values, including the AIC. On the same data, I'm comparing 3 dispersal kernels, exponential kernel (one parameter), exponential-power kernel (two-parameters) and geometric kernel (two-parameters). Is there a way, in R, to compare these dispersal kernels on the base of their AIC to find the best fitting one ? A Likelihood ratio test works only for nested models ... Thanks in advance for your help. Diane. [[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.