SAS to R translation help wanted

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

SAS to R translation help wanted

Anthony Fischbach
Satellite data collection has revolutionized our understanding of animal movmement.  
Many relocations obtained by satellite data collection suffer large inaccuracies and must be filtered.
Carla Freitas has contributed an algorithm for filtering such tracking data (http://cran.r-project.org/web/packages/argosfilter/).  For years we have used on an alternative algorithm written in SAS (http://alaska.usgs.gov/science/biology/spatial/douglas.html).  To better serve the wildlife tracking community, we wish to translate this Douglas algorithm from SAS and offer it as a function in the 'argosfilter' R package.

For this translation effort we wish to contract help.  
Interested parties, please contact me directly.
Tony Fischbach, Wildlife Biologist
Walrus Research Program
Alaska Science Center
U.S. Geological Survey
4210 University Drive
Anchorage, AK 99508-4650

AFischbach@usgs.gov
http://alaska.usgs.gov/science/biology/walrus
Reply | Threaded
Open this post in threaded view
|

Re: SAS to R translation help wanted

Michael Sumner-2
These filters are easy to implement and are available in contributed R
packages, but they are not very good. When you say "satellite"do you mean
Argos or GPS?

Argos recently updated their proprietary algorithms to give better
estimates but it it is still a rather black box.To me the issue is about
whether location accuracy of point samples or inference from derived
metrics of location are more important.


Both are very complicated and subject to active research.

I can help you implement these filters but am very loathe to if it it is
not for comparison-only purposes.

The adehabitat family of packages has very modern methods of deriving
metrics from track data.

Cheers, Mike


On Tuesday, November 8, 2011, Anthony Fischbach <[hidden email]> wrote:
> Satellite data collection has revolutionized our understanding of animal
> movmement.
> Many relocations obtained by satellite data collection suffer large
> inaccuracies and must be filtered.
> Carla Freitas has contributed an algorithm for filtering such tracking
data

> (http://cran.r-project.org/web/packages/argosfilter/).  For years we have
> used on an alternative algorithm written in SAS
> (http://alaska.usgs.gov/science/biology/spatial/douglas.html).  To better
> serve the wildlife tracking community, we wish to translate this Douglas
> algorithm from SAS and offer it as a function in the 'argosfilter' R
> package.
>
> For this translation effort we wish to contract help.
> Interested parties, please contact me directly.
>
> -----
> Tony Fischbach, Wildlife Biologist
> Walrus Research Program
> Alaska Science Center
> U.S. Geological Survey
> 4210 University Drive
> Anchorage, AK 99508-4650
>
> [hidden email]
> http://alaska.usgs.gov/science/biology/walrus
> --
> View this message in context:
http://r-sig-geo.2731867.n2.nabble.com/SAS-to-R-translation-help-wanted-tp6976543p6976543.html
> Sent from the R-sig-geo mailing list archive at Nabble.com.
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>

--
Michael Sumner
Institute for Marine and Antarctic Studies, University of Tasmania
Hobart, Australia
e-mail: [hidden email]

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: SAS to R translation help wanted

Kamran Safi-2
Hi Anthony,

if I am not mistaken, David already started to translate his filter into
R. At least I know that he wanted to do so. If you really are keen, we
could give it a try. On the other hand the filter was implemented in
www.movebank.org
Theoretically, you can upload your data to movebank, filter it according
to David's filter rules and get it back into R. The advantage being that
your data is also standardised making it easier for you to recycle your
code and you have a central repository of your data. Also you can feed
the diag files to the data base.

We are currently thinking of ways to link movebank with R, so this might
be even the better option. Let me know what you think or what would be
most helpful to wildlife trackers.

Best,

Kami



Am 09.11.2011 11:14, schrieb Michael Sumner:

> These filters are easy to implement and are available in contributed R
> packages, but they are not very good. When you say "satellite"do you mean
> Argos or GPS?
>
> Argos recently updated their proprietary algorithms to give better
> estimates but it it is still a rather black box.To me the issue is about
> whether location accuracy of point samples or inference from derived
> metrics of location are more important.
>
>
> Both are very complicated and subject to active research.
>
> I can help you implement these filters but am very loathe to if it it is
> not for comparison-only purposes.
>
> The adehabitat family of packages has very modern methods of deriving
> metrics from track data.
>
> Cheers, Mike
>
>
> On Tuesday, November 8, 2011, Anthony Fischbach<[hidden email]>  wrote:
>> Satellite data collection has revolutionized our understanding of animal
>> movmement.
>> Many relocations obtained by satellite data collection suffer large
>> inaccuracies and must be filtered.
>> Carla Freitas has contributed an algorithm for filtering such tracking
> data
>> (http://cran.r-project.org/web/packages/argosfilter/).  For years we have
>> used on an alternative algorithm written in SAS
>> (http://alaska.usgs.gov/science/biology/spatial/douglas.html).  To better
>> serve the wildlife tracking community, we wish to translate this Douglas
>> algorithm from SAS and offer it as a function in the 'argosfilter' R
>> package.
>>
>> For this translation effort we wish to contract help.
>> Interested parties, please contact me directly.
>>
>> -----
>> Tony Fischbach, Wildlife Biologist
>> Walrus Research Program
>> Alaska Science Center
>> U.S. Geological Survey
>> 4210 University Drive
>> Anchorage, AK 99508-4650
>>
>> [hidden email]
>> http://alaska.usgs.gov/science/biology/walrus
>> --
>> View this message in context:
> http://r-sig-geo.2731867.n2.nabble.com/SAS-to-R-translation-help-wanted-tp6976543p6976543.html
>> Sent from the R-sig-geo mailing list archive at Nabble.com.
>>
>> _______________________________________________
>> R-sig-Geo mailing list
>> [hidden email]
>> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>>

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo

ksafi.vcf (158 bytes) Download Attachment