Skip to content

help error: In dweibull(x, shape, scale, log) : NaNs produzidos

5 messages · Vanúcia Schumacher, David Winsemius, Peter Dalgaard

#
On Feb 21, 2012, at 9:24 AM, Van?cia Schumacher wrote:

            
It's not an error, it's only a warning, at least to the level of my  
ability to read Spanish.

It suggests you managed to send negative or infinite numbers to  
dweibull, a distribution which only supports values when given  
positive, finite numbers. Look at your data more closely.

And learn to post in plain text. HTML does a poor job of communication.
#
On Feb 21, 2012, at 16:36 , David Winsemius wrote:

            
Zeros are a common cause, too. For a < 1 (the shape parameter), the Weibull density has a singularity at zero, and fitdistr() is not tolerant of observations rounded to zero in such cases.
#
On Feb 21, 2012, at 11:43 AM, peter dalgaard wrote:

            
And that may be more correct than my guess. I see that the estimate  
for the shape parameter  from fitdistr is less than 1 and was  
surprised that dweibull(-1, 2,2) does give a returned value of 0.

 >  dweibull(-.1, 0.61473436, 2.43536064 )
[1] 0
 >  dweibull(0, 0.61473436, 2.43536064 )
[1] Inf
 >  dweibull(0, 1.61473436, 2.43536064 )
[1] 0