[R-sig-ME] about the correlated errors

zqchang at math.hkbu.edu.hk zqchang at math.hkbu.edu.hk
Sun Jun 15 13:50:50 CEST 2008

dear all:

    there are two questions,

   1.   there are several corStructs in the package,  for a given
corStruct such as

corAR1, corARMA, corCAR1, corCompSymm, corExp ...., how to obtain the

random number series with this corStruct by function as simulated

errors in a simulation procedure?

  2.  error reported :"Singularity in backsolve at level 0, block 1"

what is the possible reason and how to overcome it?





More information about the R-sig-mixed-models mailing list