Re: [GRASS-user] Univariate statistics with a moving temporal window over a STRDS
Hi Stefan, Nice problem you have on hands... you are right that min(), max(), median() are not supported in t.rast.algebra... a pity since since that would be the optimum (we could open a feature request for this as to remember). The only thing I can think of (though not elegant) is to use the where option in t.rast.aggregate setting the date and with 3 days granularity and min (max or median) method. Then cycle over starting dates... instead looking backwards, you go forward in time but iterate over each day... Then, there's Luca's add-on, t.rast.what.aggr, that allows to estimate stats backwards in time but for a vector of points with date as attribute: https://grass.osgeo.org/grass76/manuals/addons/t.rast.what.aggr.html I added Thomas Leppelt in cc, maybe he has a better idea since Sören and him were the authors of the temporal algebra (sorry for double posting Thomas, I do not know if you are in the list) best, Vero El mar., 25 feb. 2020 a las 6:02, Stefan Blumentrath (< stefan.blumentr...@nina.no>) escribió: > Dear all, > > > > I would like to calculate a running univariate statistics over a Space > Time Raster Dataset: e.g. for each day the maximum temperature looking up > to three days back. > > In t.rast.algebra, I can use temporal buffering (buff_t() function) or > temporal offset for that purpose, but functions like min(), max(), > median(), are not supported. > > t.rast.aggregate is not suitable because it does not allow for overlapping > aggregation… > > > > Any idea how to achieve that? Did I overlook a relevant module? > > > > Cheers, > > Stefan > ___ > grass-user mailing list > grass-user@lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/grass-user ___ grass-user mailing list grass-user@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Univariate statistics with a moving temporal window over a STRDS
Dear all, I would like to calculate a running univariate statistics over a Space Time Raster Dataset: e.g. for each day the maximum temperature looking up to three days back. In t.rast.algebra, I can use temporal buffering (buff_t() function) or temporal offset for that purpose, but functions like min(), max(), median(), are not supported. t.rast.aggregate is not suitable because it does not allow for overlapping aggregation... Any idea how to achieve that? Did I overlook a relevant module? Cheers, Stefan ___ grass-user mailing list grass-user@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Univariate Statistics
Greetings I need to do some univariate statitstics that goes beyond the max/min/median/average/standard deviation. I need to do some outliers detection, distribution analysis and some more basic statistics. Since GRASS doesn't perform this I would like to know what are people using? Some statistical Python module (e.g. Statpy) or other software? Thanks Best regards, Luis ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Univariate Statistics
Luis Lisboa pisze: Greetings I need to do some univariate statitstics that goes beyond the max/min/median/average/standard deviation. I need to do some outliers detection, distribution analysis and some more basic statistics. Since GRASS doesn't perform this I would like to know what are people using? Some statistical Python module (e.g. Statpy) or other software? Thanks Best regards, Luis ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user The easiest way is grass + R + spgrass6, Jarek ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Univariate Statistics
You probably should consider R. Take a look at http://grass.itc.it/statsgrass/grass_geostats.html Jose Luis Lisboa escribió: Greetings I need to do some univariate statitstics that goes beyond the max/min/median/average/standard deviation. I need to do some outliers detection, distribution analysis and some more basic statistics. Since GRASS doesn't perform this I would like to know what are people using? Some statistical Python module (e.g. Statpy) or other software? Thanks Best regards, Luis ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user -- José A. Ruiz-Arias Solar Radiation and Atmosphere Modelling Group http://matras.ujaen.es Department of Physics University of Jaén Campus Lagunillas, Building A3, Dep 066 23071, Jaén (Spain) Tlf: +34 953 212474 Fax: +34 953 212838 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Univariate Statistics
Thank you I realized that there is a r.univar.zone and an r.univar. What is the main difference between them? (they seems alike) Luis On Tue, Mar 2, 2010 at 9:33 AM, Jose A. Ruiz-Arias jarar...@ujaen.eswrote: You probably should consider R. Take a look at http://grass.itc.it/statsgrass/grass_geostats.html Jose Luis Lisboa escribió: Greetings I need to do some univariate statitstics that goes beyond the max/min/median/average/standard deviation. I need to do some outliers detection, distribution analysis and some more basic statistics. Since GRASS doesn't perform this I would like to know what are people using? Some statistical Python module (e.g. Statpy) or other software? Thanks Best regards, Luis ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user -- José A. Ruiz-Arias Solar Radiation and Atmosphere Modelling Group http://matras.ujaen.es Department of Physics University of Jaén Campus Lagunillas, Building A3, Dep 066 23071, Jaén (Spain) Tlf: +34 953 212474 Fax: +34 953 212838 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Univariate Statistics
Luis Lisboa pisze: Thank you I realized that there is a r.univar.zone and an r.univar. What is the main difference between them? (they seems alike) Luis On Tue, Mar 2, 2010 at 9:33 AM, Jose A. Ruiz-Arias jarar...@ujaen.es mailto:jarar...@ujaen.es wrote: You probably should consider R. Take a look at http://grass.itc.it/statsgrass/grass_geostats.html Jose Luis Lisboa escribió: Greetings I need to do some univariate statitstics that goes beyond the max/min/median/average/standard deviation. I need to do some outliers detection, distribution analysis and some more basic statistics. Since GRASS doesn't perform this I would like to know what are people using? Some statistical Python module (e.g. Statpy) or other software? Thanks Best regards, Luis ___ grass-user mailing list grass-user@lists.osgeo.org mailto:grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user -- José A. Ruiz-Arias Solar Radiation and Atmosphere Modelling Group http://matras.ujaen.es Department of Physics University of Jaén Campus Lagunillas, Building A3, Dep 066 23071, Jaén (Spain) Tlf: +34 953 212474 Fax: +34 953 212838 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user r.univar.zonal calculates statistics for more than one area at once. R univar only for whole map ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user