Re: [OSGeo-Discuss] Staistical analysis support needed
On Mon, Sep 13, 2010 at 03:49:46AM -0700, mayank_agarwal wrote: Hello everyone, Finally I have decided to use Open Layers on client side in conjunction with J2EE Technologies, GeoServer as a web map server and, POST GIS as Spatial Database. Now integrating all of these I want to do Statistical data analysis on the spatio-temporal data using- 1. Moran's I 2. Geary's C 3. Kriging 4. Local Moran's I 5. Spatial Scan Statistic 6. Geographic weighted regression Does anyone has any idea on how to proceed further using these combination? Is there any other client side library that I can use for this? My guess is that there are two ways that you might explore to do this. 1. Any time someone says 'statistics', my first thought is 'R'; I don't know any of the things you're talking about here, but if these are statistical methods of some sort, R seems like a valid place to start looking. http://cran.r-project.org/web/packages/rgdal/index.html rgdal will let you read OGR data sources (like PostGIS databases) into a spatial vector object, and interact with them from there. 2. I know that at least some of these mechanisms (like Kriging) are implemented in GRASS, and I would bet that this is your second best option. GRASS is a bit intimidating to a first time user (though I guess not much more so than R :)), but a very powerful geographic data analysis tool. Best of luck, -- Christopher Schmidt Web Developer ___ Discuss mailing list Discuss@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/discuss
Re: [OSGeo-Discuss] Staistical analysis support needed
Finally I have decided to use Open Layers on client side in conjunction with J2EE Technologies, GeoServer as a web map server and, POST GIS as Spatial Database. http://opengeo.org/community/ http://opengeo.org/community/suite/ Now integrating all of these I want to do Statistical data analysis on the spatio-temporal data using- 1. Moran's I 2. Geary's C 3. Kriging 4. Local Moran's I 5. Spatial Scan Statistic 6. Geographic weighted regression JRI is a Java/R Interface http://www.rforge.net/JRI/ Noli ___ Discuss mailing list Discuss@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/discuss
Re: [OSGeo-Discuss] Staistical analysis support needed
Are you building a desktop or web based application here? I see that you have selected an OpenLayers client, so that suggests web to me. In that case, since you want J2EE technologies, I would suggest Geomajas (which is fully Java - GWT). It has PostGIS support either through GeoTools or through Hibernate Spatial, and can also connect onto a Geoserver (should you still require a web map server). The analysis part would then simply be done server-side using a Java library (JRI?). mayank_agarwal schreef: Hello everyone, Finally I have decided to use Open Layers on client side in conjunction with J2EE Technologies, GeoServer as a web map server and, POST GIS as Spatial Database. Now integrating all of these I want to do Statistical data analysis on the spatio-temporal data using- 1. Moran's I 2. Geary's C 3. Kriging 4. Local Moran's I 5. Spatial Scan Statistic 6. Geographic weighted regression Does anyone has any idea on how to proceed further using these combination? Is there any other client side library that I can use for this? Thanks and regards, Mayank -- Pieter De Graef Community Manager GeoSparc nv. http://www.geosparc.com/ Chairman of the Geomajas project http://www.geomajas.org/ ___ Discuss mailing list Discuss@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/discuss
Re: [OSGeo-Discuss] Staistical analysis support needed
My guess is that there are two ways that you might explore to do this. 1. Any time someone says 'statistics', my first thought is 'R'; I don't know any of the things you're talking about here, but if these are statistical methods of some sort, R seems like a valid place to start looking. http://cran.r-project.org/web/packages/rgdal/index.html rgdal will let you read OGR data sources (like PostGIS databases) into a spatial vector object, and interact with them from there. There's also PL/R to access R within PostgreSQL, in case no one else mentioned it. HTH, Yves ___ Discuss mailing list Discuss@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/discuss