Is heteroscedasticity discussed in any paper or documentation of yours?
I am interested in testing a heteroscedastic mixed-effects model with different residual variances 𝜎_𝑗 for each experimental condition. Has any of your papers discussed that implementation? Thanks.
Comments
That does not appear to be the case, but I'll ask our expert.
E.J.
Hi sherloconan,
sorry, we do not currently support this feature in JASP. Our implementation is based on afex which builds upon lme4 that does not allow to fit models with different residual variances (as far as I know). This is however possible with nlme package with R.
Cheers,
Frantisek