[R] HELP ERROR Weibull values must be > 0

[Uwe Ligges]

On 21.02.2012 18:50, Vanúcia Schumacher wrote:
>
> Yes, my data dados[[1]] have values<=0, but  what can I do to fix thi problem/? I replace values equal to 0 for 0.00001??

If they are really 0, they are not from a weibull. If you think they 
are, there is something wrong with the data. If not, the underlying 
distribution is another one. Something we cannot know, but you should 
know what is going on with your data!

Uwe Ligges





>
> thank you
>
>
>> Date: Tue, 21 Feb 2012 18:44:34 +0100
>> From: ligges at statistik.tu-dortmund.de
>> To: vanucia-schumacher at hotmail.com
>> CC: r-help at r-project.org
>> Subject: Re: [R] HELP ERROR   Weibull values must be>  0
>>
>>
>>
>> On 21.02.2012 16:48, Vanúcia Schumacher wrote:
>>>
>>> GUYS,
>>>
>>> I NEED HELP WITH ERROR:
>>>
>>> library(MASS)
>>>> dados<-read.table("mediaRGinverno.txt",header=FALSE)
>>>> vento50<-fitdistr(dados[[1]],densfun="weibull")
>>> Erro em fitdistr(dados[[1]],
>>> densfun = "weibull") :
>>>     Weibull values must be>   0
>>> WHY RETURN THIS ERROR? WHAT CAN I DO?
>>
>>
>> Probably you have values<= 0 in dados[[1]]? Hence we are sure it is not
>> a Weibull distribution underlying your data....
>>
>> Please repair your broken shift key!
>>
>> Uwe Ligges
>>
>>
>>
>>>
>>> BEST REGARDS 		 	   		
>>> 	[[alternative HTML version deleted]]
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>