Dear Cheers

You can use SPSS and GIS 10 software to solve your problem.

Depends on the resolution of your data to some extend. I'm a big fan of R personally, but for very high resolution data you might need to combine GRASS GIS and R.

Thanks Selvam and Joanne. It was really a quick reply. Just couldn't believe how good is Research Gate! The logistic regression has been performed in R but when I want to proceed to autologistic, I kind of stuck where to start. Should I start with Moran Index Selvam?

Joanne, which package r u using in R for autologistic? My data is tarher small. The pixel resolution(grid) is 100 km x 100 km. I have 80 columns x 60 rows. However...I am dealing with monthly time series images from 1982 to 2010. I would love to try in R.

Thanks

I haven't tried autologistic regression myself, but you might find a way to start here http://r.789695.n4.nabble.com/autologistic-modelling-in-R-td874823.html. Otherwise have a look at http://stackoverflow.com, you generally get very quick replies from them.

Your data sound very similar to the kind of data I have used in my PhD. The trick is to save the data as an array, so one spatial image per array level, so you can scroll through your years. The easiest to achieve this is probably using the raster package in R.

SAM software is something easy and light to use. Is commonly used from users that work on Species Distribution Modelling.

You can find it here > http://www.ecoevol.ufg.br/sam/

Dimitris. Thank a lot! Trying to use it now.

Does anyone knows where I can find documentation about how SAM runs the auto-logistic regression? The help/ documentation link is not working. Thanks

Hi Daniela, please find the SAM Tutorial as attached. Hope it helps. :)

Thanks Idham, but I was thinking in something more detailed on the specifics of the auto-logistic regression.

Using R, the suggested autologistic.citrus() crashes using the data in the package vignette. I'm also interested in fitting a spatial model (lag, error, and/or Durbin) to a binary response variable using R. Any thoughts are appreciated.  

I have malaria incidence data recorded at health facilities spanning 15 years and would like to relate the incidences to malaria vector suitability which I converted to binary. I suppose the incidence data from year to year is temporally autocorrelated. How can run an autologistic function taking into consideration the the temporal autocorrelation

Ok thank you let me try it. Hope it works.

Regards

Spatial Efficiency metric (SPAEF) is proven to be robust when comparing two raster maps. Python and Matlab codes are available at: http://space.geus.dk/tools_products/index.html

Two papers on the metric:

https://www.hydrol-earth-syst-sci.net/22/1299/2018/hess-22-1299-2018-assets.html

https://www.geosci-model-dev-discuss.net/gmd-2017-238/