[R] Error produced by read.zoo: "bad entries"
Dimitri Liakhovitski
dimitri.liakhovitski at gmail.com
Fri Jul 23 20:07:11 CEST 2010
David/Gabor,
you helped me a lot.
Gabor - I've run table(OrigData$Month) - and it looked weird.
I went back to my file and changed the format of the date (Month) in
Excel. I have no idea what it does - but after I saved it again, it
worked.
Thanks a lot!
I'll open another thread where I'll ask how to best save dates in
Excel if one saves them as .csv.
Dimitri
On Fri, Jul 23, 2010 at 2:00 PM, David Winsemius <dwinsemius at comcast.net> wrote:
>
> On Jul 23, 2010, at 1:50 PM, Dimitri Liakhovitski wrote:
>
>> I am expecting to see the week names as row labels of z and the
>> corresponding values (like in the "monthly" example). I am pretty sure
>> - in order to get it one needs to install the latest version of zoo.
>> I've done it just a couple of days ago.
>> I am getting the error - and nothing is produced. Can it have to do
>> with the fact that I am using the newer version of zoo?
>> Again, my full code for that OrigData.csv file I sent is:
>
> Yep, updating to the current version of zoo on CRAN, zoo_1.6-4, now produces
> an error where before with the penultimate version, zoo_1.6-3, it did not.
>
> --
> David.
>>
>> OrigData<-read.csv("OrigData.csv")
>> OrigData$Month<-as.character(OrigData$Month)
>> OrigData$Month<-as.Date(OrigData$Month,"%m/%d/%y")
>> str(OrigData)
>>
>> 'data.frame': 440 obs. of 3 variables:
>> $ Brand: Factor w/ 11 levels " Plus","agrow",..: 2 2 2 2 2 2 2 2 2 2 ...
>> $ Month:Class 'Date' num [1:440] 18262 18293 18322 18353 18383 ...
>> $ Value: int NA NA NA 100 100 100 100 100 100 99 ...
>>
>> library(zoo)
>> z <- read.zoo(OrigData, index.column = 2, split = "Brand")
>>
>> Error in merge.zoo(` Plus` = c(NA, 98L, 95L, 97L, NA, 98L, 97L, 98L, NA,
>> :
>> series cannot be merged with non-unique index entries in a series
>> In addition: There were 11 warnings (use warnings() to see them)
>>
>> warnings()
>> Warning messages:
>> 1: In zoo(rval4[[i]], ix[[i]]) :
>> some methods for “zoo” objects do not work if the index entries in
>> ‘order.by’ are not unique
>> 2: In zoo(rval4[[i]], ix[[i]]) :
>> some methods for “zoo” objects do not work if the index entries in
>> ‘order.by’ are not unique
>> 3: In zoo(rval4[[i]], ix[[i]]) :
>> etc.
>>
>> But it does not give me this error for my Monthly example - even when
>> I introduce a few NAs there.
>>
>>
>> And I get this message:
>> Error in merge.zoo(` Plus` = c(NA, 98L, 95L, 97L, NA, 98L, 97L, 98L, NA,
>> :
>> series cannot be merged with non-unique index entries in a series
>> In addition: There were 11 warnings (use warnings() to see them)
>>
>>
>> On Fri, Jul 23, 2010 at 1:41 PM, David Winsemius <dwinsemius at comcast.net>
>> wrote:
>>>
>>> On Jul 23, 2010, at 1:39 PM, Dimitri Liakhovitski wrote:
>>>
>>>> Very sorry - I mistunderstood and confused split with index.column -
>>>> totally my fault.
>>>> Ok, now I've run this line:
>>>>
>>>> z <- read.zoo(OrigData, index.column = 2, split = "Brand")
>>>>
>>>> And I am getting:
>>>> Error in merge.zoo(` Plus` = c(NA, 98L, 95L, 97L, NA, 98L, 97L, 98L, NA,
>>>> :
>>>> series cannot be merged with non-unique index entries in a series
>>>> In addition: There were 11 warnings (use warnings() to see them)
>>>
>>> I got the warnings but no error:
>>>
>>>> z <- read.zoo(OrigData, split = "Brand", index.column=2)
>>>
>>> There were 11 warnings (use warnings() to see them)
>>>>
>>>> z
>>>
>>> Plus agrow chool gress Grib inKid kid omis plet pro romil
>>> [1,] NA NA NA NA NA NA NA NA NA NA NA
>>> [2,] 98 99 98 97 96 96 100 97 97 99 96
>>> [3,] 95 100 97 99 92 97 100 97 99 100 99
>>> [4,] 97 99 94 98 91 95 99 98 98 99 95
>>> [5,] NA NA NA NA NA NA NA NA NA NA NA
>>> [6,] 98 99 98 97 93 96 99 97 98 99 96
>>> [7,] 97 100 98 98 95 96 99 98 98 100 97
>>> [8,] 98 99 94 99 96 96 99 98 98 99 97
>>> [9,] NA NA NA NA NA NA NA NA NA NA NA
>>> [10,] 98 99 98 98 95 96 99 98 98 99 97
>>> [11,] 98 99 98 99 97 96 99 98 97 99 99
>>> [12,] 97 100 96 99 95 95 99 99 97 100 96
>>> [13,] 96 100 96 96 93 0 100 96 97 100 96
>>> [14,] 98 99 98 100 94 96 100 98 97 99 99
>>> [15,] 95 100 98 99 93 95 99 99 99 99 99
>>> [16,] 97 99 96 99 94 95 98 98 90 99 95
>>> [17,] 97 100 97 96 92 0 100 96 98 100 95
>>> [18,] 96 99 98 98 96 97 100 98 99 98 98
>>> [19,] 98 100 98 98 96 97 99 98 99 99 98
>>> [20,] 98 100 97 96 95 0 100 96 98 99 96
>>> [21,] 94 100 98 99 92 97 99 98 98 98 98
>>> [22,] 98 99 98 97 96 96 99 97 98 99 97
>>> [23,] 97 100 96 96 93 0 100 95 97 100 95
>>> [24,] 97 100 98 97 93 96 99 97 98 97 95
>>> [25,] 98 100 96 97 96 94 100 97 99 99 96
>>> [26,] 98 100 98 96 95 0 100 96 98 99 95
>>> [27,] 98 100 98 97 93 96 96 97 98 99 99
>>> [28,] 99 100 98 98 92 96 100 98 99 99 97
>>> [29,] 98 100 97 95 95 0 100 95 98 99 95
>>> [30,] 99 100 98 100 98 98 99 100 99 100 99
>>> [31,] 97 99 94 97 95 95 99 97 98 98 94
>>> [32,] 98 99 98 96 95 3 100 96 97 99 96
>>> [33,] 97 99 98 99 97 97 99 99 99 99 99
>>> [34,] 96 99 95 96 94 94 98 96 96 98 93
>>> [35,] 98 99 98 97 94 54 100 97 97 99 96
>>> [36,] 95 100 97 99 95 95 99 99 98 100 99
>>> [37,] 98 99 98 98 95 96 99 98 99 99 97
>>> [38,] 98 99 98 97 96 94 100 97 97 98 96
>>> [39,] 95 100 98 100 95 97 100 99 99 100 99
>>> [40,] 97 100 95 98 93 96 99 98 98 99 96
>>>
>>> Since you didn't say what was expected, I am not in a position to know if
>>> this is success.
>>>
>>>>
>>>> And under warnings() it says:
>>>> 1: In zoo(rval4[[i]], ix[[i]]) :
>>>> some methods for “zoo” objects do not work if the index entries in
>>>> ‘order.by’ are not unique
>>>>
>>>>
>>>>
>>>> On Fri, Jul 23, 2010 at 1:13 PM, David Winsemius
>>>> <dwinsemius at comcast.net>
>>>> wrote:
>>>>>
>>>>> But, but, but.... Did you read my message about the need to correctly
>>>>> specify index columns?
>>>>>
>>>>>
>>>>> The problem is that read.zoo is reading your first column as an index
>>>>> and
>>>>> it's actually the second column that should be used for that purpose.
>>>>> --
>>>>> David.
>>>>>
>>>>> On Jul 23, 2010, at 1:01 PM, Dimitri Liakhovitski wrote:
>>>>>
>>>>>> Strange, I did attach. Attaching again. Maybe the file just doesn't go
>>>>>> through?
>>>>>> I have:
>>>>>>
>>>>>> names(OrigData):
>>>>>> [1] "Brand" "Month" "Value"
>>>>>>
>>>>>> I read ?read.zoo
>>>>>> According to that index should be the column number.
>>>>>> I thought it should be split = 1 in my case - because I am splitting
>>>>>> by
>>>>>> Brand.
>>>>>> But neither split = 1 nor split =2 work.
>>>>>> And split ="Brand" does not work either. Why?
>>>>>>
>>>>>> D.
>>>>>>
>>>>>> On Fri, Jul 23, 2010 at 12:52 PM, David Winsemius
>>>>>> <dwinsemius at comcast.net> wrote:
>>>>>>>
>>>>>>> ?read.zoo
>>>>>>>
>>>>>>> You didn't specify the index column correctly.
>>>>>>> On Jul 23, 2010, at 12:36 PM, Dimitri Liakhovitski wrote:
>>>>>>>
>>>>>>>> Hello!
>>>>>>>>
>>>>>>>> I have a data set similar to the data set "monthly" in the example
>>>>>>>> below:
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> monthly<-data.frame(month=c(20090301,20090401,20090501,20100301,20100401,20090301,20090401,20090501,20100301,20100401),monthly.value=c(100,200,300,101,201,10,20,30,11,21),market=c("Market
>>>>>>>> A","Market A", "Market A","Market A", "Market A","Market B", "Market
>>>>>>>> B","Market B","Market B", "Market B"))
>>>>>>>> monthly$month<-as.character(monthly$month)
>>>>>>>> monthly$month<-as.Date(monthly$month,"%Y%m%d")
>>>>>>>> (monthly)
>>>>>>>> str(monthly)
>>>>>>>>
>>>>>>>>
>>>>>>>> I am trying to use read.zoo - like in 3 lines below:
>>>>>>>> library(zoo)
>>>>>>>> z <- read.zoo(monthly, split = "market")
>>>>>>>> (z)
>>>>>>>>
>>>>>>>> With the artificially produced data set above, it works just fine.
>>>>>>>> However, with my data it gives me an error:
>>>>>>>>
>>>>>>>> OrigData<-read.csv("OrigData.csv")
>>>>>>>> OrigData$Month<-as.character(OrigData$Month)
>>>>>>>> OrigData$Month<-as.Date(OrigData$Month,"%m/%d/%y")
>>>>>>>> str(OrigData)
>>>>>>>>
>>>>>>>> ### The result of str(OrigData) is:
>>>>>>>> 'data.frame': 440 obs. of 3 variables:
>>>>>>>> $ Brand : Factor w/ 11 levels "aBrand","bBrand",..:
>>>>>>>> Month :Class 'Date' num [1:440] 13514 13545 13573 13604,...
>>>>>>>> Value: int NA NA NA 100 100 100 100 100 100 99
>>>>>>>
>>>>>>> ?read.zoo
>>>>>>>
>>>>>>> You didn't specify the index column correctly. In this case it needs
>>>>>>> to
>>>>>>> be =
>>>>>>> 2.
>>>>>>>
>>>>>>>>
>>>>>>>> Then I try:
>>>>>>>> z <- read.zoo(OrigData, split = "Brand")
>>>>>>>>
>>>>>>>> And get the error:
>>>>>>>> Error in read.zoo(OrigData, split = "Brand") :
>>>>>>>> index has 440 bad entries at data rows: 1 2 3 4 5 6 7 8 9 10 11 12
>>>>>>>> 13
>>>>>>>>
>>>>>>>> But the structure of my OrigData is exactly the same as of monthly.
>>>>>>>> OK
>>>>>>>> - OrigData always has a few NAs in "Value" coming first - but that's
>>>>>>>> consistent for all brands.
>>>>>>>> Any idea what might be wrong?
>>>>>>>> Thanks a lot!
>>>>>>>>
>>>>>>>> Just in case -attaching the actual file.
>>>>>>>>
>>>>>>> No. Not attached.
>>>>>>>
>>>>>>> --
>>>>>>>
>>>>>>> David Winsemius, MD
>>>>>>> West Hartford, CT
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Dimitri Liakhovitski
>>>>>> Ninah Consulting
>>>>>> www.ninah.com
>>>>>> <OrigData.csv>
>>>>>
>>>>> David Winsemius, MD
>>>>> West Hartford, CT
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Dimitri Liakhovitski
>>>> Ninah Consulting
>>>> www.ninah.com
>>>
>>> David Winsemius, MD
>>> West Hartford, CT
>>>
>>>
>>
>>
>>
>> --
>> Dimitri Liakhovitski
>> Ninah Consulting
>> www.ninah.com
>
> David Winsemius, MD
> West Hartford, CT
>
>
--
Dimitri Liakhovitski
Ninah Consulting
www.ninah.com
More information about the R-help
mailing list