Experimental browser for the Atmosphere
But all too often we Bayesians fall into a trap of assuming that external data needs to be used by transforming them into a prior distribution for the new study, hiding some uncertainties and disallowing covariate adjustment. Analyze external data jointly with new data: fharrell.com/post/hxcontrol
Apr 15, 2025, 12:12 PM
{ "uri": "at://did:plc:56hsqifm2ft7ql6slwcq5tix/app.bsky.feed.post/3lmu2fabfuk2j", "cid": "bafyreib2b35ykqc6xis2smedpn45ohp5yfy6xxlzckzebrzb4gcdwpegba", "value": { "text": "But all too often we Bayesians fall into a trap of assuming that external data needs to be used by transforming them into a prior distribution for the new study, hiding some uncertainties and disallowing covariate adjustment. Analyze external data jointly with new data: fharrell.com/post/hxcontrol", "$type": "app.bsky.feed.post", "embed": { "$type": "app.bsky.embed.external", "external": { "uri": "https://fharrell.com/post/hxcontrol", "title": "Incorporating Historical Control Data Into an RCT – Statistical Thinking", "description": "Historical data (HD) are being used increasingly in Bayesian analyses when it is difficult to randomize enough patients to study effectiveness of a treatment. Such analyses summarize observational stu..." } }, "langs": [ "en" ], "reply": { "root": { "cid": "bafyreihjukegsl2ptlnfvpnc3oqrbzougbwx253s5u2gfgcaslsteyld7i", "uri": "at://did:plc:s4ahd52v7a6pyn3dlynqtoli/app.bsky.feed.post/3lmsf5rzdt22r" }, "parent": { "cid": "bafyreihjukegsl2ptlnfvpnc3oqrbzougbwx253s5u2gfgcaslsteyld7i", "uri": "at://did:plc:s4ahd52v7a6pyn3dlynqtoli/app.bsky.feed.post/3lmsf5rzdt22r" } }, "facets": [ { "index": { "byteEnd": 299, "byteStart": 272 }, "features": [ { "uri": "https://fharrell.com/post/hxcontrol", "$type": "app.bsky.richtext.facet#link" } ] } ], "createdAt": "2025-04-15T12:12:50.481Z" } }