I’m not fan of trying to select the correct model, as in real life the real model often includes effects which are too small to be well identified and then it’s not about selecting the true, but selecting something you can estimate well enough to be useful.
LOO and WAIC have high variance, which is a problem if you try to detect small differences which happens f you are comparing models which are quite similar to each other. See Comparison of Bayesian predictive methods for model selection | Statistics and Computing and GitHub - avehtari/modelselection: Tutorial on model assessment, model selection and inference after model selection. Using higher multiplier for SE, just makes it more liley that you get stuck with your baseline.
Can be prior sensitive, and not the approach with the smallest variance Comparison of Bayesian predictive methods for model selection | Statistics and Computing
There’s a problem that bias of SE depends on n and “outliers”, and thus there can’t be any “best ratio”
Sumamry: LOO (and WAIC, but since WAIC is more difficult to diagnose, I don’t recommend it) is ok for model comparison when there is a small number of comparisons. LOO (and WAIC, but…) can detect reliably only relatively big differences in the predictive performance.
After summer. This will make SE better calibrated, but it doesn’t solve the problem of relatively high SE