[R] Can R handle twin peaks - normal distribution

Spencer Graves spencer.graves at pdf.com
Thu Jun 24 16:29:23 CEST 2004


      Have you made normal probability plots?  Mixtures of a small 
number of normal distributions will appear as a collection of straight 
line segments with breakpoints related to the mixing percentages, slopes 
of the lines proportional to the standard deviations, etc.  I don't 
understand the math behind this, but some simple simulations will expose 
the patterns.  See, e.g., McLachlan and Basford (1988) Mixture Models 
(Marcel Dekker). 

      hope this helps.  spencer graves

Unung Istopo Hartanto wrote:

>Yes, I need normal mixtures plotting, because i found my data not normal
>and closer to bimodal.
>
>Thanks for All, I've installed package "nor1mix". And its solve my
>problem.
>
>regards,
>
>Unung 
>
>
>On Thu, 2004-06-24 at 16:04, Martin Maechler wrote:
>  
>
>>>>>>>"Unung" == Unung Istopo Hartanto <unung at enciety.com>
>>>>>>>    on Thu, 24 Jun 2004 10:49:29 +0700 writes:
>>>>>>>              
>>>>>>>
>>    Unung> Hi R Users, Sorry if its out of topic. I would like
>>    Unung> to ask you about twin peaks - normal
>>    Unung> distribution. How R can handle it, any example to
>>    Unung> explain it in R.
>>
>>It's not off-topic but you didn't really say what you want
>>(read the posting guide as indicated in the last line of this message!)
>>so I guess "Twin Peaks - Normal" means a mixture of two normal
>>distributions.
>>
>>There's the small 
>>package  'nor1mix'  for univariate normal mixtures plotting, RNG
>>etc, and the much more extensive package
>>'mclust' which allows to *estimate* multivariate (now including
>>uni-variate) normal mixture model parameters.
>>
>>Is it what you've wanted?
>>Martin Maechler
>>    
>>
>
>______________________________________________
>R-help at stat.math.ethz.ch mailing list
>https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
>  
>




More information about the R-help mailing list