Support multiple dimensions in stats.fit

[huard]

Addressing a Problem?

stats.fit takes a dimension string, but I think it would be useful to be able to specify a list of dimensions, to fit parameters over data from multiple dimensions (e.g. time and realization).

Potential Solution

No response

Additional context

No response

Contribution

• I would be willing/able to open a Pull Request to contribute this feature.

Code of Conduct

• I agree to follow this project's Code of Conduct

[aulemahal]

Is the solution as easy as :

daStckd = da.stack(stat_dim=['time', 'realization'])
stats.fit(daStckd)

?

If yes, could this be an example somewhere instead ?

[huard]

Yes, that's indeed the easiest way to solve this. I wonder if it's worth supporting out of the box. Maybe not.