i2_ml()

[daniel1noble]

Yefeng: "using i2_ml() to calculate confidence intervals for an intercept-only model (rma object), we need to fit intercept-only model with argument “mods = ~ 1”. Otherwise, CIs can not be returned by i2_ml"

[daniel1noble]

Need to fix this or make it more clear.

[itchyshin]

Probably we want to take bout data argument like anybody else.

I also got this error -

i2_ml(m4, data =  dat3, boot = 10)
The object/variable ('mods_formula') specified for the 'mods' argument is NULL.

I think we should probably someone automatically do this - as many people would use i2_ml without fitting the intercept as mod

[itchyshin]

Probably we need to incorporate tryCatch or something like that

"Bootstrapping [>----------------------------------------------------]   1% ETA:  1mError: Optimizer (nlminb) did not achieve convergence (convergence = 1)"

Remember to look at r2_ml as well

[daniel1noble]

Thanks @itchyshin. TryCatch should already be implemented, but clearly it's still missing something. I'll see if I can figure out where and fix that

[daniel1noble]

We have tried this again and cannot track the issues. It seems to be working with this case and many others so closing this issue