[R] glm.nb error

[Marc Schwartz]

On Jun 7, 2013, at 9:44 AM, Daofeng Li <lidaof at gmail.com> wrote:

> Dear R Community,
> 
> I have encountered a problem while using the R function glm.nb.
> The code that produce the error was following two lines:
> 
> group=c(1,1,1,1,0,0,0,0)
> fit=glm.nb(y~group)
> 
> While the y contains 8 sets of number like:
> gene275        0       1       0       0       1       5       1       0
> 
> Error message:
> 
> Error in while ((it <- it + 1) < limit && abs(del) > eps) { :
>  missing value where TRUE/FALSE needed
> Calls: glm.nb -> as.vector -> theta.ml
> In addition: There were 50 or more warnings (use warnings() to see the
> first 50)
> Execution halted
> 
> 
> Information of my system:
>> sessionInfo()
> R version 3.0.1 (2013-05-16)
> Platform: x86_64-unknown-linux-gnu (64-bit)
> 
> locale:
> [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C
> [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
> [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8
> [7] LC_PAPER=C                 LC_NAME=C
> [9] LC_ADDRESS=C               LC_TELEPHONE=C
> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
> 
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
> 
> Does anyone happen to have some hit on how to solve this?
> Appreciate for any response.
> 
> Thanks in advance,
> 
> Daofeng


There is something wrong with your actual 'y' or 'group' that is not evident from the above info:


group <- c(1, 1, 1, 1, 0, 0, 0, 0)
y <- c(0, 1, 0, 0, 1, 5, 1, 0)

> require(MASS)
Loading required package: MASS

> glm.nb(y ~ group)

Call:  glm.nb(formula = y ~ group, init.theta = 1.711564307, link = log)

Coefficients:
(Intercept)        group  
     0.5596      -1.9459  

Degrees of Freedom: 7 Total (i.e. Null);  6 Residual
Null Deviance:	    10.23 
Residual Deviance: 6.848 	AIC: 25.25


Check str(y) and str(group)

You should also be sure to note in your posts when you are using a function from a non-base package, in this case MASS, which is not indicated in your sessionInfo() above, so something is amiss there as well.

Regards,

Marc Schwartz