[R] Modeling Time Series with Missing Observations

Berend Hasselman bhh at xs4all.nl
Fri Mar 24 09:49:17 CET 2017


There is also a package imputeTS, which may be able to do what you want.
It has a nice Introduction vignette  and also appears to have nice plot functions

Berend Hasselman


> On 24 Mar 2017, at 02:11, David Winsemius <dwinsemius at comcast.net> wrote:
> 
> There's also an irts-package "irregular time series"
> 
> Sent from my iPhone
> 
>> On Mar 23, 2017, at 3:42 PM, John C Frain <frainj at gmail.com> wrote:
>> 
>> Google "arima missing data r"  will bring up several references including
>> http://stats.stackexchange.com/questions/104565/how-to-use-auto-arima-to-impute-missing-values.
>> There are several other useful results in that search.
>> 
>> John C Frain
>> 3 Aranleigh Park
>> Rathfarnham
>> Dublin 14
>> Ireland
>> www.tcd.ie/Economics/staff/frainj/home.html
>> mailto:frainj at tcd.ie
>> mailto:frainj at gmail.com
>> 
>>> On 23 March 2017 at 15:42, Jeff Newmiller <jdnewmil at dcn.davis.ca.us> wrote:
>>> 
>>> Even the most basic introduction to R discusses the use of NA for missing
>>> data. Injecting values that could be mistaken for actual readings is a
>>> dangerous approach. You can use the merge function to introduce missing
>>> rows into zoo objects or data frames.
>>> --
>>> Sent from my phone. Please excuse my brevity.
>>> 
>>> On March 23, 2017 8:22:47 AM PDT, Paul Bernal <paulbernal07 at gmail.com>
>>> wrote:
>>>> Dear all,
>>>> 
>>>> Hope you are doing well. I am trying to model the historical number of
>>>> transits of a particular market segment, but the problem is that I have
>>>> missing data.
>>>> 
>>>> I am working with monthly data, so I have 12 observations per year (in
>>>> general). The problem is that, when I bring the data from the database,
>>>> the
>>>> following happens, for example:
>>>> 
>>>> January-2000, Feb-2000, Apr-2000, Jun 2000 (I have missing
>>>> observations)
>>>> 
>>>> when I am supposed to have the sequence January-2000, Feb-2000,
>>>> Mar-2000,
>>>> Apr-2000, May-2000, Jun-2000, etc.
>>>> 
>>>> How can I model a time series when there are missing months? I was
>>>> planning
>>>> making up fictional or fake observations with a value of 1 to fill in
>>>> the
>>>> gaps but not sure if this is a reasonable approach.
>>>> 
>>>> Any help and/or guidance will be greatly appreciated,
>>>> 
>>>> Best regards,
>>>> 
>>>> Paul
>>>> 
>>>>     [[alternative HTML version deleted]]
>>>> 
>>>> ______________________________________________
>>>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>>>> 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 -- To UNSUBSCRIBE and more, see
>>> 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.
>>> 
>> 
>>   [[alternative HTML version deleted]]
>> 
>> ______________________________________________
>> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
>> 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 -- To UNSUBSCRIBE and more, see
> 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.



More information about the R-help mailing list