[R] Boxplot philosophy {was "Boxplot in R"}

[(Ted Harding)]

On 12-Jul-05 Ted Harding wrote:
> On 12-Jul-05 Spencer Graves wrote:
>>         I'll bite:  How does one detect bimodalidty from a boxplot?
>> 
>>         spencer graves
>> 
>> Berton Gunter wrote:
>> [...]
>>> Finally, a tidbit for boxplot afficianados: how does one detect
>>> bimodality from a boxplot?
>>> 
>>> -- Bert Gunter
> 
> Not definitively detectable of course, but certainly suspectable.
> 
> Hint: What is the relationship between the distances from the "median"
> line to the extremes of the "box", and from the extremes of the "box"
> to the ends of the "whiskers", in the case of a uniform distribution?
> 
> 
> 
> If that doesn't trigger it, then:
> 
> Hint:
> 
>   unimod<-rnorm(1000,0,sqrt(26))
>   nomod<-sqrt(12*26)*runif(1000)
>   boxplot(data.frame(cbind(unimod,nomod,bimod)))

OOPS!! (cut-&-pasted the wrong lines):

  unimod<-rnorm(1000,0,sqrt(26))
  nomod<-sqrt(12*26)*(runif(1000)-0.5)
  bimod<-c(rnorm(500,-5,1),rnorm(500,5,1))
  boxplot(data.frame(cbind(unimod,nomod,bimod)))

> Best wishes,
> Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 094 0861
Date: 12-Jul-05                                       Time: 10:24:34
------------------------------ XFMail ------------------------------