[R] Setting breaks to data more appropriately

[David Winsemius]

On Mar 23, 2010, at 5:23 PM, ROLL Josh F wrote:

> David
> Thanks you for the guidance.  I am looking to drive the data into  
> the intervals.  Problem now is that I have a few 0 values that are  
> returning as NA (e.g -> (0.1,0.2] (0.1,0.2] (0.1,0.2] <NA>       
> (0,0.1]  )
> This is causing issues when I try and use the brks to inform a  
> clorepleth map.  I tried setting all na's to 0 but because the  
> dataset is in factor form (I guess this is why) it wont allow me to  
> do so. Thoughts about how to handle the zero values?  Thanks
> JR
>

?cut

By default include.lowest is set to FALSE. Setting it to TRUE should  
cure your missing zeroes problem. (My own feeling is that was a very  
poor choice of defaults but I guess it is cast in stone now.)

You don't say how you tried setting NA's to zero (and doing so is not  
a good idea if there are real zeroes in the data), but the is.na and  
is.na<- functions are often needed for that purpose.

-- 
David.
> -----Original Message-----
> From: David Winsemius [mailto:dwinsemius at comcast.net]
> Sent: Monday, March 22, 2010 12:46 PM
> To: ROLL Josh F
> Cc: r-help at r-project.org
> Subject: Re: [R] Setting breaks to data more appropriately
>
>
> On Mar 22, 2010, at 1:49 PM, LCOG1 wrote:
>
>>
>> Basic question.  For the below data, i would like to but each of the
>> values in a bin that represents their value.  So the below would
>> hopefully put .1 in the 0-.1 bin, .2 in the .11-.2 bin and so forth.
>> The outlying values would then be put into and outer category
>> representing everything
>>> 1.  Im
>> using the breaks to inform some code for making a clorepleth map that
>> represents probabilities, which in some cases IS greater than 1
>
> ... not if it's a quantile or a probability.
>
>> and i need
>> to identify those better.
>
> Define "better".
>
>> As my code stands now, my real data is put put into this form when
>> brks is called:
>>
>>       0%        10%        20%        30%        40%
>> 50%        60%
>> 0.00000000 0.05054675 0.07787235 0.11235238 0.14424786 0.18089360
>> 0.21475990
>>      70%        80%        90%       100%
>> 0.26309899 0.30807771 0.39478573 0.67573483.
>>
>> But what i want is for the values to be placed in bins corresponding
>> to their value(0-.1, .11-.2, .21-.3 etc)
>>
>> Pct.SFD<-c(.1,.2,.3,.4,.5,.6,.7,.8,.9,1,2,3)
>> brks <- quantile(Pct.SFD, )
>>
>> I think this is clear.
>
> It's not. You need to decide whether you want the breaking to be  
> driven by you or by the data. If you are doing the driving then use
>
> cut(object, breaks=c(seq(0,1, by=0.1), Inf) , right=TRUE)
>
> If the data is doing the driving then:
>
> cut(object, breaks=quantile(object, probs= seq(0,1,1/10 ) ) ,
> right=TRUE)
>
> --
> David.
>
>> Thanks
>> --
>> View this message in context:
>> http://n4.nabble.com/Setting-breaks-to-data-more-appropriately- 
>> tp16780
>> 19p1678019.html Sent from the R help mailing list archive at
>> Nabble.com.
>>
>> ______________________________________________
>> 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.
>
>

David Winsemius, MD
West Hartford, CT