1.8 06 November 2014
   Fixed an error where mu<0 was flagged as an error, even for the Gaussian case.
   The code now checks that the dispersion is greater than zero, the original intent.
   Thanks for Lars Ronnegaard and Gordon Smyth for identifying this error.
   
1.7  30 October 2012
	Minor edits to code and help files.

1.6  15/Dec/2008
	Fixed a minor error in dglm.control.Rd

1.6  15/Dec/2008
  Dispersion model was incorrectly fitted (even in the examples)
   compared to the given specification; introduced a fix that is
   not entirely satisfactory, but works.


1.5   06 May 2006
  Corrected a reference to the dlink as dlink$name to name.dlink

   
1.4:
  Changed four occurrences of  F  rather than  FALSE  in
  print.summary.dglm.

1.3:  2006/11/28
  
  Obviously didn't fix the 1.2 error; hopefully now fixed.
  This was accomplished by adding a new input, etastart=NULL.

  

1.2: 2006/09/25

   Fixed an error identified by Sonja Kuhnt for Gaussian models:
   Error in eval(expr, envir, enclos) : object "etastart" not found

   

v1.1: 

   Fixed an error identified by Cristiano Amancio for binomial data:

   Error in while (abs(m2loglikold - m2loglik)/(abs(m2loglikold) + 1) > epsilon &&  :
           absent value where TRUE/FALSE necessary

   Furthermore: Warning message:
   NaNs produced in: log(x) ".


