{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreid2g5r55ojitzz55im5c7ngxqzfxfk5kxqov7hdpsiwuczzmsae2m",
    "uri": "at://did:plc:wwyqal4cnqhuwyacdj7rqq3n/app.bsky.feed.post/3mj3rdaz5uoc2"
  },
  "path": "/t/power-calculations-in-longitudinal-mixed-effects-from-two-measurements-to-three-measurements/28699#post_4",
  "publishedAt": "2026-04-09T19:21:18.000Z",
  "site": "https://discourse.datamethods.org",
  "textContent": "That’s excellent Marc - a nice addition to `lme4`. Adding random effects to AR(1) will help meet correlation structure assumptions, and having AR(1) in the model will make the random effects smaller which adds stability. In a similar vein I’ve seen one example where there was lack of fit of a Markov-1 ordinal model until random intercepts were added.\n\nWhen there are random effects, it is more natural to use Bayesian models, which handle them better than trying to approximate marginal sampling distributions.",
  "title": "Power Calculations in Longitudinal Mixed Effects - from two measurements to three measurements"
}