Andrea Manica

I just had a look at the commented out code for a loglik test, and it is easy to fix it to give the same results as in R (the...

I made a PR #109 .The proposed implementation matches the lme4 `anova.merMod` in behaviour. It's a bit tricky to write tests. The ideal scenario would be to run `anova.merMod` with...

No worries, give me a couple of days and I'll add some tests to the PR.

@ejolly Actually, I had a bit of spare time, so tests are now fully implemented. Whilst I was at it, I also added a wrapper to `confint` for the Lmer...

I don't think it's a GDAL version issue. If I use: ``` sf::gdal_utils( util = "buildvrt", source = c(file_name_1,file_name_2), destination = vrt_filename, options = c("-separate","-overwrite") ) ``` I can create...

This is a bit of "bandaid" solution, but I can prevent the warnings when reading back the file by helping GDAL find the time axis with: ``` nc_in

I can confirm that, on my Ubuntu22.04, setting gdal warnings to 1 gives me the same warnings with `terra` alone. However, it has not impact of the value of "GDAL_NETCDF_IGNORE_XY_AXIS_NAME_CHECKS",...

According to my reading of the CF standards, I think the current time units are not quite right, as the CF standard requires a date as a reference point even...

Years and months are valid units, but tricky, in the CF standards, which is probably why gdal does not consider them. They don't really conform to the use of year...

Thank you @simonpcouch for looking into this. I have just checked, and the situation is the same in the original analysis that inspired the `reprex`, so at least this is...