[R] Table rearranging

[David Winsemius]

On Feb 10, 2012, at 4:47 PM, Jeffrey Joh wrote:

>
> Hi David,
> Totally forgot line 3!  So "513 red" should be ok but "420 red" and  
> "917 yellow" aren't.
>

I don't have a plyr solution but this is a base solution:

dat[ as.logical( ave(dat$door, dat$date,
                FUN=function(x) {"open" %in% x & "closed" %in% x} )) , ]
   ID measurement date   door color
1  1   0.9352938  513   open   red
3  3   0.9620535  513 closed   red
4  4   0.9639099 1230   open  blue
5  5   0.9765203 1230   open green
6  6   0.9893108 1230 closed  blue
8  8   0.8945757 1230   open  blue

(You could have used Dunlap's resutl and tested for both the meanOpen  
and meanClosed values being not-is.na.)

-- 
David

> Jeffrey
>
> ----------------------------------------
>> CC: r-help at r-project.org
>> From: dwinsemius at comcast.net
>> To: johjeffrey at hotmail.com
>> Subject: Re: [R] Table rearranging
>> Date: Tue, 7 Feb 2012 21:51:55 -0500
>>
>>
>> On Feb 7, 2012, at 8:43 PM, Jeffrey Joh wrote:
>>
>>>
>>> Hi David, I am not sure how ddply/summarize solves my issue. I have
>>> the following table:
>>>
>>> ID measurement date door color
>>> 1 0.93529385 513 open red
>>> 2 0.97419293 420 open red
>>> 3 0.962053514 513 closed red
>>> 4 0.963909937 1230 open blue
>>> 5 0.97652034 1230 open green
>>> 6 0.989310795 1230 closed blue
>>> 7 0.9941022 917 closed yellow
>>> 8 0.8945757 1230 open blue
>>>
>>> I only want to keep the lines that have corresponding open/closed
>>> measurements. For example, I want to keep lines 4,6,8 because for
>>> the "1230 blue" condition, there exists both open and closed
>>> measurements.
>>>
>>> However, the "513 red" condition has an open measurement, but no
>>> closed measurement.
>>
>> Huh? what about line 3?
>>
>> --
>> David,
>>> Therefore, line 1 should be deleted.
>>>
>>> Jeffrey
>>>
>>>
>>> ----------------------------------------
>>>> CC: r-help at r-project.org; wdunlap at tibco.com
>>>> From: dwinsemius at comcast.net
>>>> To: johjeffrey at hotmail.com
>>>> Subject: Re: [R] Table rearranging
>>>> Date: Tue, 7 Feb 2012 09:08:00 -0500
>>>>
>>>>
>>>> On Feb 7, 2012, at 4:21 AM, Jeffrey Joh wrote:
>>>>
>>>>>
>>>>> Thank you for your help, Bill.
>>>>>
>>>>>> From the original table (not the plyr output), I would like to
>>>>>> remove all the lines that do not have a corresponding open/closed
>>>>>> measurement. For example, if there is a Closed yellow measurement
>>>>>> on 0917, but not an Open yellow 0917 measurement, then the Closed
>>>>>> yellow should be deleted.
>>>>>
>>>>> How can I make this change?
>>>>>
>>>>
>>>> In R you need to assign the results of a function to an object name
>>>> so
>>>> you code would look like:
>>>>
>>>> modified_data <- ddply(d, .(date, color), summarize,
>>>> meanClosed=mean(measurement[door=="closed"]),
>>>> nClosed=sum(door=="closed"))
>>>>
>>>> --
>>>> David
>>>>> Jeffrey
>>>>>
>>>>>
>>>>> ----------------------------------------
>>>>>> From: wdunlap at tibco.com
>>>>>
>>>>>> To: johjeffrey at hotmail.com; r-help at r-project.org
>>>>>
>>>>>> Subject: RE: [R] Table rearranging
>>>>>
>>>>>> Date: Tue, 7 Feb 2012 00:43:25 +0000
>>>>>
>>>>>>
>>>>>
>>>>>> Install and load the "plyr" package and try something like:
>>>>>
>>>>>>
>>>>>
>>>>>>> ddply(d, .(date, color), summarize,
>>>>>
>>>>>> + ddply(d, .(date, color), summarize
>>>>>
>>>>>> + meanClosed=mean(measurement[door=="closed"]),
>>>>>> nClosed=sum(door=="closed"))
>>>>>
>>>>>> date color meanOpen nOpen meanClosed nClosed
>>>>>
>>>>>> 1 420 red 0.9741929 1 NaN 0
>>>>>
>>>>>> 2 513 red 0.9352938 1 0.9620535 1
>>>>>
>>>>>> 3 917 yellow NaN 0 0.9941022 1
>>>>>
>>>>>> 4 1230 blue 0.9639099 1 0.9893108 1
>>>>>
>>>>>> 5 1230 green 0.9765203 1 NaN 0
>>>>>
>>>>>>
>>>>>
>>>>>> Bill Dunlap
>>>>>
>>>>>> Spotfire, TIBCO Software
>>>>>
>>>>>> wdunlap tibco.com
>>>>>
>>>>>>
>>>>>
>>>>>>> -----Original Message-----
>>>>>
>>>>>>> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org
>>>>>>> ] On Behalf Of Jeffrey Joh
>>>>>
>>>>>>> Sent: Monday, February 06, 2012 4:28 PM
>>>>>
>>>>>>> To: r-help at r-project.org
>>>>>
>>>>>>> Subject: [R] Table rearranging
>>>>>
>>>>>>>
>>>>>
>>>>>>>
>>>>>
>>>>>>> I have a table that looks like this:
>>>>>
>>>>>>>
>>>>>
>>>>>>> measurement date door color
>>>>>
>>>>>>> 0.93529385 513 open red
>>>>>
>>>>>>> 0.97419293 420 open red
>>>>>
>>>>>>> 0.962053514 513 closed red
>>>>>
>>>>>>> 0.963909937 1230 open blue
>>>>>
>>>>>>> 0.97652034 1230 open green
>>>>>
>>>>>>> 0.989310795 1230 closed blue
>>>>>
>>>>>>> 0.9941022 917 closed yellow
>>>>>
>>>>>>>
>>>>>
>>>>>>> I would like to create a table that has: Open measurement,  
>>>>>>> Closed
>>>>>>> measurement, date, color. For every
>>>>>
>>>>>>> date/color combination, there should be two columns to represent
>>>>>>> the door open/closed measurement.
>>>>>
>>>>>>>
>>>>>
>>>>>>> If there are multiple datapoints with a given door/date/color
>>>>>>> combination, then they should be
>>>>>
>>>>>>> averaged.
>>>>>
>>>>>>> I would also like to make two columns to represent the number of
>>>>>
>>>>>>> datapoints that were averaged in determining the open/closed
>>>>>
>>>>>>> measurements.
>>>>>
>>>>>>>
>>>>>
>>>>>>> Jeffrey
>>>>>
>>>>>>>
>>>>>
>>>>>>> ______________________________________________
>>>>>
>>>>>>> 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.
>>>>>
>>>>>
>>>>> ______________________________________________
>>>>> 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
>>>>
>>>
>>
>> David Winsemius, MD
>> West Hartford, CT
>>
> 		 	   		

David Winsemius, MD
West Hartford, CT