[R-sig-Geo] Spatial prediction map using raster format dataset

[Omar Faisel]

Actually, I am a new user in R environment. and i faced few dilemmas, if you could help me with some tips regarding: My data represent landslides locations and conditioning factors contain the following: I have spatial data for landslide assessment in .tiff format.
1)Dependent: Training and testing (1=landslide, 0= no landslide, -9999 value refer to nodata) 2)Independent variables, slope, elevation..etc.My question: should i keep using raster format and search for suitable analysis methods like regressions . OR, i need to convert them to ascii, and use the huge amount of tools available. Note: if i convert my data into ascii, it would develop uncertainty, moreover, ascii file size will be huge (60 MB).Worth to mention that my training data represents polygon converted to raster. I am totally overwhelmed. kindly any suggestions. 
Thank you @Eelke . Actually the mentioned case (Vignette) deals with points feature. My question about when we deal with such feature but in raster format. Any suggestions? thank you again

Cordially,
0mar. 


     On Thursday, October 15, 2015 1:19 PM, Eelke Folmer <e.o.folmer at gmail.com> wrote:
   
 

 Omar,
I think that the following document will be of help:
https://cran.r-project.org/web/packages/dismo/vignettes/sdm.pdf
It is a gentle and practical introduction to raster: both data handling 
and regression. It is about species distribution modeling but the 
principles apply to your question too, I think.
Eelke

On 10/15/2015 07:43 AM, Omar Faisel wrote:
>
>
>
> Dear All,
> My data represent landslides locations and conditioning factors (5000 row, 2000 column, tiff format)  contain the following:1- Training and testing locations (1=ist, 0= not, also -9999 value refer to no-data areas)2- Independent variables like slope, elevation..etc.
> Which type of code do i need to use :1- To read the files, (which format?)2- To run linear regression or any prediction model.
> Thank you in advance.
>
> Omar.
>
>
>
>
>
>     [[alternative HTML version deleted]]
>
>

_______________________________________________
R-sig-Geo mailing list
R-sig-Geo at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-geo



  
	[[alternative HTML version deleted]]