[R] function censReg in panel data setting

Arne Henningsen arne.henningsen at googlemail.com
Wed Sep 7 13:44:27 CEST 2011

Dear Igors

On 7 September 2011 10:39, Igors <igors.lahanciks at gmail.com> wrote:
> However I am still struggling to obtain model estmates.
> The same code:
>> UpC <- censReg(Power ~ Windspeed, left = -Inf, right =
>> 2000,data=PData_In,method="BHHH",nGHQ = 4,start=c(-691.18,186.79,3.9,3.9))
> Error in maxNRCompute(fn = logLikAttr, fnOrig = fn, gradOrig = grad,
> hessOrig = hess,  :
>  NA in the initial gradient
> I have tried to change starting values and regressors in the model several
> times, but I always get the same mentioned error message. How can I make it
> work?  Is this function maxNRCompute() on the last step of calculation
> (maximization of the ML)?
> I had an idea that the error could appear since I have huge sample, but I
> tried to cut it and it still doesn't work.

It is hard to figure out the cause of this error without a
reproducible example. Is is possible that you send a reproducible
example to me?

Could it be that there are NAs in the data or something in the panel
data specification is not as censReg() expects it?


Arne Henningsen

More information about the R-help mailing list