[R] function censReg in panel data setting

Arne Henningsen arne.henningsen at googlemail.com
Tue Sep 6 11:14:56 CEST 2011

On 6 September 2011 07:51, Igors <igors.lahanciks at gmail.com> wrote:
>>You chose a suitable solution for the first problem (NA in initial
>>gradient). Unfortunately, the documentation of "censReg" is not very
>>clear regarding starting values of panel tobit models. Please note
>>that you have to specify 4 starting values: intercept, slope
>>parameter, variance of the individual effects ("mu"), and variance of
>>the general error term ("nu").
> I have already tried to use 4 parameters, here is what I get:
>> UpC <- censReg(Power ~ Windspeed, left = -Inf, right =
>> 2000,data=PData_In,method="BHHH",nGHQ = 4,start=c(-691,189,5,5))
> Error in censReg(Power ~ Windspeed, left = -Inf, right = 2000, data =
> PData_In,  :
>  argument 'start' must have length 3
> Any ideas how to overcome this one? Could this be an error in package or
> censReg function?

Yes, you are right. This is (was) a bug in the censReg function/package.

I have fixed the package. The source code of the new version (0.5-6)
is available on R-Forge [1]; R packages will be available on CRAN [2]
and R-Forge [3] probably within one or two days.

[1] https://r-forge.r-project.org/scm/?group_id=256
[2] http://cran.r-project.org/package=censReg
[3] https://r-forge.r-project.org/R/?group_id=256


Arne Henningsen

More information about the R-help mailing list